diff --git a/providers/tempr/logo.svg b/providers/tempr/logo.svg new file mode 100644 index 00000000000..a9fb18e6892 --- /dev/null +++ b/providers/tempr/logo.svg @@ -0,0 +1,4 @@ + + + + diff --git a/providers/tempr/models/aws-bedrock/amazon.nova-2-lite-v1:0.toml b/providers/tempr/models/aws-bedrock/amazon.nova-2-lite-v1:0.toml new file mode 100644 index 00000000000..ef0dca1df3c --- /dev/null +++ b/providers/tempr/models/aws-bedrock/amazon.nova-2-lite-v1:0.toml @@ -0,0 +1,20 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high +# /v1/responses: reasoning.effort = none|low|medium|high +base_model = "amazon/nova-2-lite" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high"] }, +] + +[cost] +input = 0.33 +output = 2.75 +cache_read = 0.0825 +cache_write = 0.33 + +[limit] +output = 65_535 diff --git a/providers/tempr/models/aws-bedrock/amazon.nova-lite-v1:0.toml b/providers/tempr/models/aws-bedrock/amazon.nova-lite-v1:0.toml new file mode 100644 index 00000000000..d14e250d61f --- /dev/null +++ b/providers/tempr/models/aws-bedrock/amazon.nova-lite-v1:0.toml @@ -0,0 +1,7 @@ +base_model = "amazon/nova-lite" + +[cost] +input = 0.06 +output = 0.24 +cache_read = 0.015 +cache_write = 0.06 diff --git a/providers/tempr/models/aws-bedrock/amazon.nova-micro-v1:0.toml b/providers/tempr/models/aws-bedrock/amazon.nova-micro-v1:0.toml new file mode 100644 index 00000000000..b87c4d866cc --- /dev/null +++ b/providers/tempr/models/aws-bedrock/amazon.nova-micro-v1:0.toml @@ -0,0 +1,7 @@ +base_model = "amazon/nova-micro" + +[cost] +input = 0.035 +output = 0.14 +cache_read = 0.00875 +cache_write = 0.035 diff --git a/providers/tempr/models/aws-bedrock/amazon.nova-pro-v1:0.toml b/providers/tempr/models/aws-bedrock/amazon.nova-pro-v1:0.toml new file mode 100644 index 00000000000..7b3f1664bf1 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/amazon.nova-pro-v1:0.toml @@ -0,0 +1,7 @@ +base_model = "amazon/nova-pro" + +[cost] +input = 0.80 +output = 3.20 +cache_read = 0.20 +cache_write = 0.80 diff --git a/providers/tempr/models/aws-bedrock/anthropic.claude-fable-5-1.toml b/providers/tempr/models/aws-bedrock/anthropic.claude-fable-5-1.toml new file mode 100644 index 00000000000..5a89c5cec9e --- /dev/null +++ b/providers/tempr/models/aws-bedrock/anthropic.claude-fable-5-1.toml @@ -0,0 +1,15 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "anthropic/claude-fable-5-1" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 10 +output = 50 +cache_read = 0.25 +cache_write = 12.5 diff --git a/providers/tempr/models/aws-bedrock/anthropic.claude-fable-5.toml b/providers/tempr/models/aws-bedrock/anthropic.claude-fable-5.toml new file mode 100644 index 00000000000..9261d534a9b --- /dev/null +++ b/providers/tempr/models/aws-bedrock/anthropic.claude-fable-5.toml @@ -0,0 +1,15 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "anthropic/claude-fable-5" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 10 +output = 50 +cache_read = 1 +cache_write = 12.5 diff --git a/providers/tempr/models/aws-bedrock/anthropic.claude-haiku-4-5-20251001-v1:0.toml b/providers/tempr/models/aws-bedrock/anthropic.claude-haiku-4-5-20251001-v1:0.toml new file mode 100644 index 00000000000..1d465c9a75e --- /dev/null +++ b/providers/tempr/models/aws-bedrock/anthropic.claude-haiku-4-5-20251001-v1:0.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-haiku-4-5-20251001" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 1.00 +output = 5.00 +cache_read = 0.10 +cache_write = 1.25 diff --git a/providers/tempr/models/aws-bedrock/anthropic.claude-opus-4-1-20250805-v1:0.toml b/providers/tempr/models/aws-bedrock/anthropic.claude-opus-4-1-20250805-v1:0.toml new file mode 100644 index 00000000000..cc71d06a39d --- /dev/null +++ b/providers/tempr/models/aws-bedrock/anthropic.claude-opus-4-1-20250805-v1:0.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-opus-4-1-20250805" +status = "deprecated" + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 15 +output = 75 +cache_read = 1.5 +cache_write = 18.75 diff --git a/providers/tempr/models/aws-bedrock/anthropic.claude-opus-4-5-20251101-v1:0.toml b/providers/tempr/models/aws-bedrock/anthropic.claude-opus-4-5-20251101-v1:0.toml new file mode 100644 index 00000000000..34045eb287a --- /dev/null +++ b/providers/tempr/models/aws-bedrock/anthropic.claude-opus-4-5-20251101-v1:0.toml @@ -0,0 +1,20 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high (alias: top-level reasoning_effort) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high +base_model = "anthropic/claude-opus-4-5-20251101" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high"] }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 5.00 +output = 25.00 +cache_read = 0.50 +cache_write = 6.25 diff --git a/providers/tempr/models/aws-bedrock/anthropic.claude-opus-4-6-v1.toml b/providers/tempr/models/aws-bedrock/anthropic.claude-opus-4-6-v1.toml new file mode 100644 index 00000000000..e4b6522ee56 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/anthropic.claude-opus-4-6-v1.toml @@ -0,0 +1,20 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|max (alias: top-level reasoning_effort) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|max; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high|max +base_model = "anthropic/claude-opus-4-6" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "max"] }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 5.50 +output = 27.50 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/anthropic.claude-opus-4-7.toml b/providers/tempr/models/aws-bedrock/anthropic.claude-opus-4-7.toml new file mode 100644 index 00000000000..e95807e726b --- /dev/null +++ b/providers/tempr/models/aws-bedrock/anthropic.claude-opus-4-7.toml @@ -0,0 +1,17 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-4-7" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5.00 +output = 25.00 +cache_read = 0.50 +cache_write = 6.25 diff --git a/providers/tempr/models/aws-bedrock/anthropic.claude-opus-4-8.toml b/providers/tempr/models/aws-bedrock/anthropic.claude-opus-4-8.toml new file mode 100644 index 00000000000..0f9c2a17e48 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/anthropic.claude-opus-4-8.toml @@ -0,0 +1,17 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-4-8" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5 +output = 25 +cache_read = 0.5 +cache_write = 6.25 diff --git a/providers/tempr/models/aws-bedrock/anthropic.claude-opus-5.toml b/providers/tempr/models/aws-bedrock/anthropic.claude-opus-5.toml new file mode 100644 index 00000000000..5e542272cc4 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/anthropic.claude-opus-5.toml @@ -0,0 +1,15 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-5" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5 +output = 25 +cache_read = 0.5 +cache_write = 6.25 diff --git a/providers/tempr/models/aws-bedrock/anthropic.claude-sonnet-4-5-20250929-v1:0.toml b/providers/tempr/models/aws-bedrock/anthropic.claude-sonnet-4-5-20250929-v1:0.toml new file mode 100644 index 00000000000..7fb292bc1ef --- /dev/null +++ b/providers/tempr/models/aws-bedrock/anthropic.claude-sonnet-4-5-20250929-v1:0.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-sonnet-4-5-20250929" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3.00 +output = 15.00 +cache_read = 0.30 +cache_write = 3.75 diff --git a/providers/tempr/models/aws-bedrock/anthropic.claude-sonnet-4-6.toml b/providers/tempr/models/aws-bedrock/anthropic.claude-sonnet-4-6.toml new file mode 100644 index 00000000000..8f7b96e84a0 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/anthropic.claude-sonnet-4-6.toml @@ -0,0 +1,23 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|max (alias: top-level reasoning_effort) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|max; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high|max +base_model = "anthropic/claude-sonnet-4-6" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "max"] }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3.30 +output = 16.50 +cache_read = 0.33 +cache_write = 4.125 + +[limit] +output = 128_000 diff --git a/providers/tempr/models/aws-bedrock/anthropic.claude-sonnet-5.toml b/providers/tempr/models/aws-bedrock/anthropic.claude-sonnet-5.toml new file mode 100644 index 00000000000..2dfc5026ed8 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/anthropic.claude-sonnet-5.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-sonnet-5" +structured_output = false + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 2 +output = 10 +cache_read = 0.2 +cache_write = 2.5 diff --git a/providers/tempr/models/aws-bedrock/apac.amazon.nova-lite-v1:0.toml b/providers/tempr/models/aws-bedrock/apac.amazon.nova-lite-v1:0.toml new file mode 100644 index 00000000000..fea8d067992 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/apac.amazon.nova-lite-v1:0.toml @@ -0,0 +1,8 @@ +base_model = "amazon/nova-lite" +name = "Nova Lite (APAC)" + +[cost] +input = 0.063 +output = 0.252 +cache_read = 0.01575 +cache_write = 0.063 diff --git a/providers/tempr/models/aws-bedrock/apac.amazon.nova-micro-v1:0.toml b/providers/tempr/models/aws-bedrock/apac.amazon.nova-micro-v1:0.toml new file mode 100644 index 00000000000..b90a49c4146 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/apac.amazon.nova-micro-v1:0.toml @@ -0,0 +1,8 @@ +base_model = "amazon/nova-micro" +name = "Nova Micro (APAC)" + +[cost] +input = 0.037 +output = 0.148 +cache_read = 0.00925 +cache_write = 0.037 diff --git a/providers/tempr/models/aws-bedrock/apac.amazon.nova-pro-v1:0.toml b/providers/tempr/models/aws-bedrock/apac.amazon.nova-pro-v1:0.toml new file mode 100644 index 00000000000..3afa4954356 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/apac.amazon.nova-pro-v1:0.toml @@ -0,0 +1,8 @@ +base_model = "amazon/nova-pro" +name = "Nova Pro (APAC)" + +[cost] +input = 0.840 +output = 3.360 +cache_read = 0.210 +cache_write = 0.840 diff --git a/providers/tempr/models/aws-bedrock/apac.anthropic.claude-sonnet-4-20250514-v1:0.toml b/providers/tempr/models/aws-bedrock/apac.anthropic.claude-sonnet-4-20250514-v1:0.toml new file mode 100644 index 00000000000..4dcaa1297b4 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/apac.anthropic.claude-sonnet-4-20250514-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-sonnet-4-20250514" +name = "Claude Sonnet 4 (APAC)" +status = "deprecated" + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3 +output = 15 +cache_read = 0.3 +cache_write = 3.75 diff --git a/providers/tempr/models/aws-bedrock/au.anthropic.claude-haiku-4-5-20251001-v1:0.toml b/providers/tempr/models/aws-bedrock/au.anthropic.claude-haiku-4-5-20251001-v1:0.toml new file mode 100644 index 00000000000..f6b829fda03 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/au.anthropic.claude-haiku-4-5-20251001-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-haiku-4-5-20251001" +name = "Claude Haiku 4.5 (AU)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 1.10 +output = 5.50 +cache_read = 0.11 +cache_write = 1.375 diff --git a/providers/tempr/models/aws-bedrock/au.anthropic.claude-opus-4-6-v1.toml b/providers/tempr/models/aws-bedrock/au.anthropic.claude-opus-4-6-v1.toml new file mode 100644 index 00000000000..ee4f5924efb --- /dev/null +++ b/providers/tempr/models/aws-bedrock/au.anthropic.claude-opus-4-6-v1.toml @@ -0,0 +1,21 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|max (alias: top-level reasoning_effort) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|max; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high|max +base_model = "anthropic/claude-opus-4-6" +name = "AU Anthropic Claude Opus 4.6" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "max"] }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 5.50 +output = 27.50 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/au.anthropic.claude-opus-4-7.toml b/providers/tempr/models/aws-bedrock/au.anthropic.claude-opus-4-7.toml new file mode 100644 index 00000000000..e09265e1fc9 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/au.anthropic.claude-opus-4-7.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-4-7" +name = "Claude Opus 4.7 (AU)" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5.50 +output = 27.50 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/au.anthropic.claude-opus-4-8.toml b/providers/tempr/models/aws-bedrock/au.anthropic.claude-opus-4-8.toml new file mode 100644 index 00000000000..c27fb13a401 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/au.anthropic.claude-opus-4-8.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-4-8" +name = "Claude Opus 4.8 (AU)" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5.50 +output = 27.50 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/au.anthropic.claude-opus-5.toml b/providers/tempr/models/aws-bedrock/au.anthropic.claude-opus-5.toml new file mode 100644 index 00000000000..01d55bb31bb --- /dev/null +++ b/providers/tempr/models/aws-bedrock/au.anthropic.claude-opus-5.toml @@ -0,0 +1,16 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-5" +name = "Claude Opus 5 (AU)" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5.50 +output = 27.50 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/au.anthropic.claude-sonnet-4-5-20250929-v1:0.toml b/providers/tempr/models/aws-bedrock/au.anthropic.claude-sonnet-4-5-20250929-v1:0.toml new file mode 100644 index 00000000000..ae3e8dc2d39 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/au.anthropic.claude-sonnet-4-5-20250929-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-sonnet-4-5-20250929" +name = "Claude Sonnet 4.5 (AU)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3.30 +output = 16.50 +cache_read = 0.33 +cache_write = 4.125 diff --git a/providers/tempr/models/aws-bedrock/au.anthropic.claude-sonnet-4-6.toml b/providers/tempr/models/aws-bedrock/au.anthropic.claude-sonnet-4-6.toml new file mode 100644 index 00000000000..9456529b3d0 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/au.anthropic.claude-sonnet-4-6.toml @@ -0,0 +1,24 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|max (alias: top-level reasoning_effort) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|max; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high|max +base_model = "anthropic/claude-sonnet-4-6" +name = "AU Anthropic Claude Sonnet 4.6" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "max"] }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3.30 +output = 16.50 +cache_read = 0.33 +cache_write = 4.125 + +[limit] +output = 128_000 diff --git a/providers/tempr/models/aws-bedrock/au.anthropic.claude-sonnet-5.toml b/providers/tempr/models/aws-bedrock/au.anthropic.claude-sonnet-5.toml new file mode 100644 index 00000000000..22f739fb304 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/au.anthropic.claude-sonnet-5.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-sonnet-5" +name = "Claude Sonnet 5 (AU)" +structured_output = false + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 2.20 +output = 11 +cache_read = 0.22 +cache_write = 2.75 diff --git a/providers/tempr/models/aws-bedrock/ca.amazon.nova-lite-v1:0.toml b/providers/tempr/models/aws-bedrock/ca.amazon.nova-lite-v1:0.toml new file mode 100644 index 00000000000..92ffc99f6d6 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/ca.amazon.nova-lite-v1:0.toml @@ -0,0 +1,8 @@ +base_model = "amazon/nova-lite" +name = "Nova Lite (CA)" + +[cost] +input = 0.064 +output = 0.256 +cache_read = 0.016 +cache_write = 0.064 diff --git a/providers/tempr/models/aws-bedrock/deepseek.r1-v1:0.toml b/providers/tempr/models/aws-bedrock/deepseek.r1-v1:0.toml new file mode 100644 index 00000000000..8d6650b7b00 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/deepseek.r1-v1:0.toml @@ -0,0 +1,11 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# No reasoning control for this model on this host: it reasons at its own depth, +# and the x-tempr-reasoning-effort response header reports "default". +base_model = "deepseek/deepseek-r1" +tool_call = false + +reasoning_options = [] + +[cost] +input = 1.35 +output = 5.4 diff --git a/providers/tempr/models/aws-bedrock/deepseek.v3-v1:0.toml b/providers/tempr/models/aws-bedrock/deepseek.v3-v1:0.toml new file mode 100644 index 00000000000..8db724b2e98 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/deepseek.v3-v1:0.toml @@ -0,0 +1,20 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# /v1/messages: thinking.type = enabled|disabled +# /v1/responses: reasoning.effort = "none" turns it off, any other level turns it on +base_model = "deepseek/deepseek-v3.1" +last_updated = "2025-09-18" +interleaved = true +structured_output = true + +reasoning_options = [ + { type = "toggle" }, +] + +[cost] +input = 0.58 +output = 1.68 + +[limit] +context = 163_840 +output = 81_920 diff --git a/providers/tempr/models/aws-bedrock/deepseek.v3.2.toml b/providers/tempr/models/aws-bedrock/deepseek.v3.2.toml new file mode 100644 index 00000000000..e74aff3eaa7 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/deepseek.v3.2.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# /v1/messages: thinking.type = enabled|disabled +# /v1/responses: reasoning.effort = "none" turns it off, any other level turns it on +base_model = "deepseek/deepseek-v3.2" +last_updated = "2026-02-06" +interleaved = true + +reasoning_options = [ + { type = "toggle" }, +] + +[cost] +input = 0.62 +output = 1.85 + +[limit] +context = 163_840 +output = 81_920 diff --git a/providers/tempr/models/aws-bedrock/eu.amazon.nova-2-lite-v1:0.toml b/providers/tempr/models/aws-bedrock/eu.amazon.nova-2-lite-v1:0.toml new file mode 100644 index 00000000000..75d61187da4 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.amazon.nova-2-lite-v1:0.toml @@ -0,0 +1,21 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high +# /v1/responses: reasoning.effort = none|low|medium|high +base_model = "amazon/nova-2-lite" +name = "Nova 2 Lite (EU)" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high"] }, +] + +[cost] +input = 0.374 +output = 3.157 +cache_read = 0.0935 +cache_write = 0.374 + +[limit] +output = 65_535 diff --git a/providers/tempr/models/aws-bedrock/eu.amazon.nova-lite-v1:0.toml b/providers/tempr/models/aws-bedrock/eu.amazon.nova-lite-v1:0.toml new file mode 100644 index 00000000000..cce6a1ab142 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.amazon.nova-lite-v1:0.toml @@ -0,0 +1,8 @@ +base_model = "amazon/nova-lite" +name = "Nova Lite (EU)" + +[cost] +input = 0.069 +output = 0.276 +cache_read = 0.01725 +cache_write = 0.069 diff --git a/providers/tempr/models/aws-bedrock/eu.amazon.nova-micro-v1:0.toml b/providers/tempr/models/aws-bedrock/eu.amazon.nova-micro-v1:0.toml new file mode 100644 index 00000000000..f1f89fcd868 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.amazon.nova-micro-v1:0.toml @@ -0,0 +1,8 @@ +base_model = "amazon/nova-micro" +name = "Nova Micro (EU)" + +[cost] +input = 0.040 +output = 0.160 +cache_read = 0.010 +cache_write = 0.040 diff --git a/providers/tempr/models/aws-bedrock/eu.amazon.nova-pro-v1:0.toml b/providers/tempr/models/aws-bedrock/eu.amazon.nova-pro-v1:0.toml new file mode 100644 index 00000000000..07c3f65655b --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.amazon.nova-pro-v1:0.toml @@ -0,0 +1,8 @@ +base_model = "amazon/nova-pro" +name = "Nova Pro (EU)" + +[cost] +input = 0.920 +output = 3.680 +cache_read = 0.230 +cache_write = 0.920 diff --git a/providers/tempr/models/aws-bedrock/eu.anthropic.claude-fable-5.toml b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-fable-5.toml new file mode 100644 index 00000000000..9d42ecead12 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-fable-5.toml @@ -0,0 +1,16 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "anthropic/claude-fable-5" +name = "Claude Fable 5 (EU)" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 11 +output = 55 +cache_read = 1.1 +cache_write = 13.75 diff --git a/providers/tempr/models/aws-bedrock/eu.anthropic.claude-haiku-4-5-20251001-v1:0.toml b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-haiku-4-5-20251001-v1:0.toml new file mode 100644 index 00000000000..613d7561efd --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-haiku-4-5-20251001-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-haiku-4-5-20251001" +name = "Claude Haiku 4.5 (EU)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 1.10 +output = 5.50 +cache_read = 0.11 +cache_write = 1.375 diff --git a/providers/tempr/models/aws-bedrock/eu.anthropic.claude-opus-4-5-20251101-v1:0.toml b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-opus-4-5-20251101-v1:0.toml new file mode 100644 index 00000000000..956da69b084 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-opus-4-5-20251101-v1:0.toml @@ -0,0 +1,21 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high (alias: top-level reasoning_effort) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high +base_model = "anthropic/claude-opus-4-5-20251101" +name = "Claude Opus 4.5 (EU)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high"] }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 5.50 +output = 27.50 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/eu.anthropic.claude-opus-4-6-v1.toml b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-opus-4-6-v1.toml new file mode 100644 index 00000000000..626cf49a9a5 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-opus-4-6-v1.toml @@ -0,0 +1,21 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|max (alias: top-level reasoning_effort) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|max; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high|max +base_model = "anthropic/claude-opus-4-6" +name = "Claude Opus 4.6 (EU)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "max"] }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 5.5 +output = 27.5 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/eu.anthropic.claude-opus-4-7.toml b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-opus-4-7.toml new file mode 100644 index 00000000000..d4678975d75 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-opus-4-7.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-4-7" +name = "Claude Opus 4.7 (EU)" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5.50 +output = 27.50 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/eu.anthropic.claude-opus-4-8.toml b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-opus-4-8.toml new file mode 100644 index 00000000000..652ac13566e --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-opus-4-8.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-4-8" +name = "Claude Opus 4.8 (EU)" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5.5 +output = 27.5 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/eu.anthropic.claude-opus-5.toml b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-opus-5.toml new file mode 100644 index 00000000000..119ee08936b --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-opus-5.toml @@ -0,0 +1,16 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-5" +name = "Claude Opus 5 (EU)" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5.5 +output = 27.5 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/eu.anthropic.claude-sonnet-4-20250514-v1:0.toml b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-sonnet-4-20250514-v1:0.toml new file mode 100644 index 00000000000..ed65aad82c7 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-sonnet-4-20250514-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-sonnet-4-20250514" +name = "Claude Sonnet 4 (EU)" +status = "deprecated" + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3 +output = 15 +cache_read = 0.3 +cache_write = 3.75 diff --git a/providers/tempr/models/aws-bedrock/eu.anthropic.claude-sonnet-4-5-20250929-v1:0.toml b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-sonnet-4-5-20250929-v1:0.toml new file mode 100644 index 00000000000..d18fa87cf3c --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-sonnet-4-5-20250929-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-sonnet-4-5-20250929" +name = "Claude Sonnet 4.5 (EU)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3.30 +output = 16.50 +cache_read = 0.33 +cache_write = 4.125 diff --git a/providers/tempr/models/aws-bedrock/eu.anthropic.claude-sonnet-4-6.toml b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-sonnet-4-6.toml new file mode 100644 index 00000000000..afae620d2a7 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-sonnet-4-6.toml @@ -0,0 +1,24 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|max (alias: top-level reasoning_effort) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|max; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high|max +base_model = "anthropic/claude-sonnet-4-6" +name = "Claude Sonnet 4.6 (EU)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "max"] }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3.3 +output = 16.5 +cache_read = 0.33 +cache_write = 4.125 + +[limit] +output = 128_000 diff --git a/providers/tempr/models/aws-bedrock/eu.anthropic.claude-sonnet-5.toml b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-sonnet-5.toml new file mode 100644 index 00000000000..ae848222ef0 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.anthropic.claude-sonnet-5.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-sonnet-5" +name = "Claude Sonnet 5 (EU)" +structured_output = false + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 2.2 +output = 11 +cache_read = 0.22 +cache_write = 2.75 diff --git a/providers/tempr/models/aws-bedrock/eu.mistral.pixtral-large-2502-v1:0.toml b/providers/tempr/models/aws-bedrock/eu.mistral.pixtral-large-2502-v1:0.toml new file mode 100644 index 00000000000..4558cfda30b --- /dev/null +++ b/providers/tempr/models/aws-bedrock/eu.mistral.pixtral-large-2502-v1:0.toml @@ -0,0 +1,6 @@ +base_model = "mistral/pixtral-large-2502" +name = "Pixtral Large (25.02) (EU)" + +[cost] +input = 2.00 +output = 6.00 diff --git a/providers/tempr/models/aws-bedrock/global.amazon.nova-2-lite-v1:0.toml b/providers/tempr/models/aws-bedrock/global.amazon.nova-2-lite-v1:0.toml new file mode 100644 index 00000000000..99fffffaf4d --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.amazon.nova-2-lite-v1:0.toml @@ -0,0 +1,21 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high +# /v1/responses: reasoning.effort = none|low|medium|high +base_model = "amazon/nova-2-lite" +name = "Nova 2 Lite (Global)" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high"] }, +] + +[cost] +input = 0.300 +output = 2.500 +cache_read = 0.075 +cache_write = 0.300 + +[limit] +output = 65_535 diff --git a/providers/tempr/models/aws-bedrock/global.anthropic.claude-fable-5-1.toml b/providers/tempr/models/aws-bedrock/global.anthropic.claude-fable-5-1.toml new file mode 100644 index 00000000000..98d56033fb8 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.anthropic.claude-fable-5-1.toml @@ -0,0 +1,16 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "anthropic/claude-fable-5-1" +name = "Claude Fable 5.1 (Global)" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 10 +output = 50 +cache_read = 0.25 +cache_write = 12.5 diff --git a/providers/tempr/models/aws-bedrock/global.anthropic.claude-fable-5.toml b/providers/tempr/models/aws-bedrock/global.anthropic.claude-fable-5.toml new file mode 100644 index 00000000000..e7b02074f3b --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.anthropic.claude-fable-5.toml @@ -0,0 +1,16 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "anthropic/claude-fable-5" +name = "Claude Fable 5 (Global)" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 10 +output = 50 +cache_read = 1 +cache_write = 12.5 diff --git a/providers/tempr/models/aws-bedrock/global.anthropic.claude-haiku-4-5-20251001-v1:0.toml b/providers/tempr/models/aws-bedrock/global.anthropic.claude-haiku-4-5-20251001-v1:0.toml new file mode 100644 index 00000000000..2deaa67140a --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.anthropic.claude-haiku-4-5-20251001-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-haiku-4-5-20251001" +name = "Claude Haiku 4.5 (Global)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 1.00 +output = 5.00 +cache_read = 0.10 +cache_write = 1.25 diff --git a/providers/tempr/models/aws-bedrock/global.anthropic.claude-opus-4-5-20251101-v1:0.toml b/providers/tempr/models/aws-bedrock/global.anthropic.claude-opus-4-5-20251101-v1:0.toml new file mode 100644 index 00000000000..5c1cf1befb2 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.anthropic.claude-opus-4-5-20251101-v1:0.toml @@ -0,0 +1,21 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high (alias: top-level reasoning_effort) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high +base_model = "anthropic/claude-opus-4-5-20251101" +name = "Claude Opus 4.5 (Global)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high"] }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 5.00 +output = 25.00 +cache_read = 0.50 +cache_write = 6.25 diff --git a/providers/tempr/models/aws-bedrock/global.anthropic.claude-opus-4-6-v1.toml b/providers/tempr/models/aws-bedrock/global.anthropic.claude-opus-4-6-v1.toml new file mode 100644 index 00000000000..3a20f6f45b1 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.anthropic.claude-opus-4-6-v1.toml @@ -0,0 +1,21 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|max (alias: top-level reasoning_effort) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|max; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high|max +base_model = "anthropic/claude-opus-4-6" +name = "Claude Opus 4.6 (Global)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "max"] }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 5 +output = 25 +cache_read = 0.5 +cache_write = 6.25 diff --git a/providers/tempr/models/aws-bedrock/global.anthropic.claude-opus-4-7.toml b/providers/tempr/models/aws-bedrock/global.anthropic.claude-opus-4-7.toml new file mode 100644 index 00000000000..33f9f348238 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.anthropic.claude-opus-4-7.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-4-7" +name = "Claude Opus 4.7 (Global)" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5.00 +output = 25.00 +cache_read = 0.50 +cache_write = 6.25 diff --git a/providers/tempr/models/aws-bedrock/global.anthropic.claude-opus-4-8.toml b/providers/tempr/models/aws-bedrock/global.anthropic.claude-opus-4-8.toml new file mode 100644 index 00000000000..e520cc2bbaa --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.anthropic.claude-opus-4-8.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-4-8" +name = "Claude Opus 4.8 (Global)" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5 +output = 25 +cache_read = 0.5 +cache_write = 6.25 diff --git a/providers/tempr/models/aws-bedrock/global.anthropic.claude-opus-5.toml b/providers/tempr/models/aws-bedrock/global.anthropic.claude-opus-5.toml new file mode 100644 index 00000000000..26cfa30b23b --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.anthropic.claude-opus-5.toml @@ -0,0 +1,16 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-5" +name = "Claude Opus 5 (Global)" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5 +output = 25 +cache_read = 0.5 +cache_write = 6.25 diff --git a/providers/tempr/models/aws-bedrock/global.anthropic.claude-sonnet-4-20250514-v1:0.toml b/providers/tempr/models/aws-bedrock/global.anthropic.claude-sonnet-4-20250514-v1:0.toml new file mode 100644 index 00000000000..a4e81ba0809 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.anthropic.claude-sonnet-4-20250514-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-sonnet-4-20250514" +name = "Claude Sonnet 4 (Global)" +status = "deprecated" + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3 +output = 15 +cache_read = 0.3 +cache_write = 3.75 diff --git a/providers/tempr/models/aws-bedrock/global.anthropic.claude-sonnet-4-5-20250929-v1:0.toml b/providers/tempr/models/aws-bedrock/global.anthropic.claude-sonnet-4-5-20250929-v1:0.toml new file mode 100644 index 00000000000..36de66e005a --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.anthropic.claude-sonnet-4-5-20250929-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-sonnet-4-5-20250929" +name = "Claude Sonnet 4.5 (Global)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3.00 +output = 15.00 +cache_read = 0.30 +cache_write = 3.75 diff --git a/providers/tempr/models/aws-bedrock/global.anthropic.claude-sonnet-4-6.toml b/providers/tempr/models/aws-bedrock/global.anthropic.claude-sonnet-4-6.toml new file mode 100644 index 00000000000..1f4ce9f540f --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.anthropic.claude-sonnet-4-6.toml @@ -0,0 +1,24 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|max (alias: top-level reasoning_effort) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|max; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high|max +base_model = "anthropic/claude-sonnet-4-6" +name = "Claude Sonnet 4.6 (Global)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "max"] }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3 +output = 15 +cache_read = 0.3 +cache_write = 3.75 + +[limit] +output = 128_000 diff --git a/providers/tempr/models/aws-bedrock/global.anthropic.claude-sonnet-5.toml b/providers/tempr/models/aws-bedrock/global.anthropic.claude-sonnet-5.toml new file mode 100644 index 00000000000..02cd7553be7 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.anthropic.claude-sonnet-5.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-sonnet-5" +name = "Claude Sonnet 5 (Global)" +structured_output = false + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 2 +output = 10 +cache_read = 0.2 +cache_write = 2.5 diff --git a/providers/tempr/models/aws-bedrock/global.openai.gpt-5.6-luna.toml b/providers/tempr/models/aws-bedrock/global.openai.gpt-5.6-luna.toml new file mode 100644 index 00000000000..afea644c5ef --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.openai.gpt-5.6-luna.toml @@ -0,0 +1,28 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "openai/gpt-5.6-luna" +base_model_omit = ["cost.context_over_200k", "experimental.modes.fast"] +name = "GPT-5.6 Luna (Global)" + +reasoning_options = [ + { type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 0.20 +output = 1.20 +cache_read = 0.02 +cache_write = 0.25 + +[[cost.tiers]] +tier = { size = 272_000 } +input = 0.40 +output = 1.80 +cache_read = 0.04 +cache_write = 0.50 + +[modalities] +input = ["text", "image"] diff --git a/providers/tempr/models/aws-bedrock/global.openai.gpt-5.6-sol.toml b/providers/tempr/models/aws-bedrock/global.openai.gpt-5.6-sol.toml new file mode 100644 index 00000000000..8aa63ef68ec --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.openai.gpt-5.6-sol.toml @@ -0,0 +1,29 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "openai/gpt-5.6-sol" +base_model_omit = ["cost.context_over_200k", "experimental.modes.fast"] +name = "GPT-5.6 Sol (Global)" +structured_output = false + +reasoning_options = [ + { type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 4.00 +output = 20.00 +cache_read = 0.40 +cache_write = 5.00 + +[[cost.tiers]] +tier = { size = 272_000 } +input = 8.00 +output = 30.00 +cache_read = 0.80 +cache_write = 10.00 + +[modalities] +input = ["text", "image"] diff --git a/providers/tempr/models/aws-bedrock/global.openai.gpt-5.6-terra.toml b/providers/tempr/models/aws-bedrock/global.openai.gpt-5.6-terra.toml new file mode 100644 index 00000000000..d824a3b1a5a --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.openai.gpt-5.6-terra.toml @@ -0,0 +1,29 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "openai/gpt-5.6-terra" +base_model_omit = ["cost.context_over_200k", "experimental.modes.fast"] +name = "GPT-5.6 Terra (Global)" +structured_output = false + +reasoning_options = [ + { type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 2.00 +output = 12.00 +cache_read = 0.20 +cache_write = 2.50 + +[[cost.tiers]] +tier = { size = 272_000 } +input = 4.00 +output = 18.00 +cache_read = 0.40 +cache_write = 5.00 + +[modalities] +input = ["text", "image"] diff --git a/providers/tempr/models/aws-bedrock/global.openai.gpt-6-astra.toml b/providers/tempr/models/aws-bedrock/global.openai.gpt-6-astra.toml new file mode 100644 index 00000000000..0831e50c19f --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.openai.gpt-6-astra.toml @@ -0,0 +1,27 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "openai/gpt-6-astra" +base_model_omit = ["cost.context_over_200k", "experimental.modes.fast"] +name = "GPT-6 Astra (Global)" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 10.00 +output = 50.00 +cache_read = 1.00 +cache_write = 12.50 + +[[cost.tiers]] +tier = { size = 272_000 } +input = 20.00 +output = 75.00 +cache_read = 2.00 +cache_write = 25.00 + +[modalities] +input = ["text", "image"] diff --git a/providers/tempr/models/aws-bedrock/global.xai.grok-4.6.toml b/providers/tempr/models/aws-bedrock/global.xai.grok-4.6.toml new file mode 100644 index 00000000000..05547a282d7 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/global.xai.grok-4.6.toml @@ -0,0 +1,17 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh +# /v1/responses: reasoning.effort = low|medium|high|xhigh +base_model = "xai/grok-4.6" +name = "Grok 4.6 (Global)" +last_updated = "2026-08-18" +structured_output = false + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh"] }, +] + +[cost] +input = 2.00 +output = 6.00 +cache_read = 0.50 diff --git a/providers/tempr/models/aws-bedrock/google.gemma-3-12b-it.toml b/providers/tempr/models/aws-bedrock/google.gemma-3-12b-it.toml new file mode 100644 index 00000000000..5fafeb345f5 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/google.gemma-3-12b-it.toml @@ -0,0 +1,10 @@ +base_model = "google/gemma-3-12b-it" +tool_call = false +structured_output = true + +[cost] +input = 0.09 +output = 0.29 + +[limit] +output = 8_192 diff --git a/providers/tempr/models/aws-bedrock/google.gemma-3-27b-it.toml b/providers/tempr/models/aws-bedrock/google.gemma-3-27b-it.toml new file mode 100644 index 00000000000..07a67419576 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/google.gemma-3-27b-it.toml @@ -0,0 +1,11 @@ +base_model = "google/gemma-3-27b-it" +tool_call = false +structured_output = true + +[cost] +input = 0.23 +output = 0.38 + +[limit] +context = 202_752 +output = 8_192 diff --git a/providers/tempr/models/aws-bedrock/google.gemma-3-4b-it.toml b/providers/tempr/models/aws-bedrock/google.gemma-3-4b-it.toml new file mode 100644 index 00000000000..099a6ec9c43 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/google.gemma-3-4b-it.toml @@ -0,0 +1,9 @@ +base_model = "google/gemma-3-4b-it" +tool_call = false + +[cost] +input = 0.04 +output = 0.08 + +[limit] +output = 4_096 diff --git a/providers/tempr/models/aws-bedrock/google.gemma-4-26b-a4b.toml b/providers/tempr/models/aws-bedrock/google.gemma-4-26b-a4b.toml new file mode 100644 index 00000000000..a71a95eb455 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/google.gemma-4-26b-a4b.toml @@ -0,0 +1,17 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|high (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = high +# /v1/responses: reasoning.effort = none|high +base_model = "google/gemma-4-26b-a4b-it" + +reasoning_options = [ + { type = "effort", values = ["none", "high"] }, +] + +[cost] +input = 0.13 +output = 0.40 + +[modalities] +input = ["text", "image", "video"] diff --git a/providers/tempr/models/aws-bedrock/google.gemma-4-31b.toml b/providers/tempr/models/aws-bedrock/google.gemma-4-31b.toml new file mode 100644 index 00000000000..4815f8bfee8 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/google.gemma-4-31b.toml @@ -0,0 +1,17 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|high (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = high +# /v1/responses: reasoning.effort = none|high +base_model = "google/gemma-4-31b-it" + +reasoning_options = [ + { type = "effort", values = ["none", "high"] }, +] + +[cost] +input = 0.14 +output = 0.40 + +[modalities] +input = ["text", "image", "video"] diff --git a/providers/tempr/models/aws-bedrock/google.gemma-4-e2b.toml b/providers/tempr/models/aws-bedrock/google.gemma-4-e2b.toml new file mode 100644 index 00000000000..3cc4685b69e --- /dev/null +++ b/providers/tempr/models/aws-bedrock/google.gemma-4-e2b.toml @@ -0,0 +1,17 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|high (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = high +# /v1/responses: reasoning.effort = none|high +base_model = "google/gemma-4-E2B-it" + +reasoning_options = [ + { type = "effort", values = ["none", "high"] }, +] + +[cost] +input = 0.04 +output = 0.08 + +[modalities] +input = ["text", "image", "audio", "video"] diff --git a/providers/tempr/models/aws-bedrock/in.openai.gpt-5.6-luna.toml b/providers/tempr/models/aws-bedrock/in.openai.gpt-5.6-luna.toml new file mode 100644 index 00000000000..44363aee457 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/in.openai.gpt-5.6-luna.toml @@ -0,0 +1,24 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "openai/gpt-5.6-luna" +name = "GPT-5.6 Luna (India)" + +reasoning_options = [ + { type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 0.22 +output = 1.32 +cache_read = 0.022 +cache_write = 0.275 + +[[cost.tiers]] +tier = { type = "context", size = 272_000 } +input = 0.44 +output = 1.98 +cache_read = 0.044 +cache_write = 0.55 diff --git a/providers/tempr/models/aws-bedrock/in.openai.gpt-5.6-terra.toml b/providers/tempr/models/aws-bedrock/in.openai.gpt-5.6-terra.toml new file mode 100644 index 00000000000..288f06cabec --- /dev/null +++ b/providers/tempr/models/aws-bedrock/in.openai.gpt-5.6-terra.toml @@ -0,0 +1,25 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "openai/gpt-5.6-terra" +name = "GPT-5.6 Terra (India)" +structured_output = false + +reasoning_options = [ + { type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 2.20 +output = 13.20 +cache_read = 0.22 +cache_write = 2.75 + +[[cost.tiers]] +tier = { type = "context", size = 272_000 } +input = 4.40 +output = 19.80 +cache_read = 0.44 +cache_write = 5.50 diff --git a/providers/tempr/models/aws-bedrock/jp.amazon.nova-2-lite-v1:0.toml b/providers/tempr/models/aws-bedrock/jp.amazon.nova-2-lite-v1:0.toml new file mode 100644 index 00000000000..4cbfa24dac6 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/jp.amazon.nova-2-lite-v1:0.toml @@ -0,0 +1,21 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high +# /v1/responses: reasoning.effort = none|low|medium|high +base_model = "amazon/nova-2-lite" +name = "Nova 2 Lite (JP)" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high"] }, +] + +[cost] +input = 0.396 +output = 3.311 +cache_read = 0.099 +cache_write = 0.396 + +[limit] +output = 65_535 diff --git a/providers/tempr/models/aws-bedrock/jp.anthropic.claude-haiku-4-5-20251001-v1:0.toml b/providers/tempr/models/aws-bedrock/jp.anthropic.claude-haiku-4-5-20251001-v1:0.toml new file mode 100644 index 00000000000..69f374c989d --- /dev/null +++ b/providers/tempr/models/aws-bedrock/jp.anthropic.claude-haiku-4-5-20251001-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-haiku-4-5-20251001" +name = "Claude Haiku 4.5 (JP)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 1.10 +output = 5.50 +cache_read = 0.11 +cache_write = 1.375 diff --git a/providers/tempr/models/aws-bedrock/jp.anthropic.claude-opus-4-7.toml b/providers/tempr/models/aws-bedrock/jp.anthropic.claude-opus-4-7.toml new file mode 100644 index 00000000000..2e91fffa9cf --- /dev/null +++ b/providers/tempr/models/aws-bedrock/jp.anthropic.claude-opus-4-7.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-4-7" +name = "Claude Opus 4.7 (JP)" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5.50 +output = 27.50 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/jp.anthropic.claude-opus-4-8.toml b/providers/tempr/models/aws-bedrock/jp.anthropic.claude-opus-4-8.toml new file mode 100644 index 00000000000..a623fc3f9f7 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/jp.anthropic.claude-opus-4-8.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-4-8" +name = "Claude Opus 4.8 (JP)" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5.50 +output = 27.50 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/jp.anthropic.claude-opus-5.toml b/providers/tempr/models/aws-bedrock/jp.anthropic.claude-opus-5.toml new file mode 100644 index 00000000000..b7b5028d123 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/jp.anthropic.claude-opus-5.toml @@ -0,0 +1,16 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-5" +name = "Claude Opus 5 (JP)" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5.5 +output = 27.5 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/jp.anthropic.claude-sonnet-4-5-20250929-v1:0.toml b/providers/tempr/models/aws-bedrock/jp.anthropic.claude-sonnet-4-5-20250929-v1:0.toml new file mode 100644 index 00000000000..584953121ef --- /dev/null +++ b/providers/tempr/models/aws-bedrock/jp.anthropic.claude-sonnet-4-5-20250929-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-sonnet-4-5-20250929" +name = "Claude Sonnet 4.5 (JP)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3.30 +output = 16.50 +cache_read = 0.33 +cache_write = 4.125 diff --git a/providers/tempr/models/aws-bedrock/jp.anthropic.claude-sonnet-4-6.toml b/providers/tempr/models/aws-bedrock/jp.anthropic.claude-sonnet-4-6.toml new file mode 100644 index 00000000000..464e760613a --- /dev/null +++ b/providers/tempr/models/aws-bedrock/jp.anthropic.claude-sonnet-4-6.toml @@ -0,0 +1,24 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|max (alias: top-level reasoning_effort) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|max; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high|max +base_model = "anthropic/claude-sonnet-4-6" +name = "Claude Sonnet 4.6 (JP)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "max"] }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3.30 +output = 16.50 +cache_read = 0.33 +cache_write = 4.125 + +[limit] +output = 128_000 diff --git a/providers/tempr/models/aws-bedrock/jp.anthropic.claude-sonnet-5.toml b/providers/tempr/models/aws-bedrock/jp.anthropic.claude-sonnet-5.toml new file mode 100644 index 00000000000..ad857b0c68f --- /dev/null +++ b/providers/tempr/models/aws-bedrock/jp.anthropic.claude-sonnet-5.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-sonnet-5" +name = "Claude Sonnet 5 (JP)" +structured_output = false + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 2.2 +output = 11 +cache_read = 0.22 +cache_write = 2.75 diff --git a/providers/tempr/models/aws-bedrock/meta.llama3-1-70b-instruct-v1:0.toml b/providers/tempr/models/aws-bedrock/meta.llama3-1-70b-instruct-v1:0.toml new file mode 100644 index 00000000000..ecb05b56443 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/meta.llama3-1-70b-instruct-v1:0.toml @@ -0,0 +1,6 @@ +base_model = "meta/llama-3.1-70b-instruct" +name = "Llama 3.1 70B Instruct" + +[cost] +input = 0.72 +output = 0.72 diff --git a/providers/tempr/models/aws-bedrock/meta.llama3-1-8b-instruct-v1:0.toml b/providers/tempr/models/aws-bedrock/meta.llama3-1-8b-instruct-v1:0.toml new file mode 100644 index 00000000000..566da15a27d --- /dev/null +++ b/providers/tempr/models/aws-bedrock/meta.llama3-1-8b-instruct-v1:0.toml @@ -0,0 +1,6 @@ +base_model = "meta/llama-3.1-8b-instruct" +name = "Llama 3.1 8B Instruct" + +[cost] +input = 0.22 +output = 0.22 diff --git a/providers/tempr/models/aws-bedrock/meta.llama3-3-70b-instruct-v1:0.toml b/providers/tempr/models/aws-bedrock/meta.llama3-3-70b-instruct-v1:0.toml new file mode 100644 index 00000000000..4fdc5db7b82 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/meta.llama3-3-70b-instruct-v1:0.toml @@ -0,0 +1,6 @@ +base_model = "meta/llama-3.3-70b-instruct" +name = "Llama 3.3 70B Instruct" + +[cost] +input = 0.72 +output = 0.72 diff --git a/providers/tempr/models/aws-bedrock/meta.llama4-maverick-17b-instruct-v1:0.toml b/providers/tempr/models/aws-bedrock/meta.llama4-maverick-17b-instruct-v1:0.toml new file mode 100644 index 00000000000..4f79423dc2f --- /dev/null +++ b/providers/tempr/models/aws-bedrock/meta.llama4-maverick-17b-instruct-v1:0.toml @@ -0,0 +1,8 @@ +base_model = "meta/llama-4-maverick-17b-instruct" + +[cost] +input = 0.24 +output = 0.97 + +[limit] +output = 8_192 diff --git a/providers/tempr/models/aws-bedrock/meta.llama4-scout-17b-instruct-v1:0.toml b/providers/tempr/models/aws-bedrock/meta.llama4-scout-17b-instruct-v1:0.toml new file mode 100644 index 00000000000..a76d58147cc --- /dev/null +++ b/providers/tempr/models/aws-bedrock/meta.llama4-scout-17b-instruct-v1:0.toml @@ -0,0 +1,8 @@ +base_model = "meta/llama-4-scout-17b-instruct" + +[cost] +input = 0.17 +output = 0.66 + +[limit] +output = 8_192 diff --git a/providers/tempr/models/aws-bedrock/minimax.minimax-m2.1.toml b/providers/tempr/models/aws-bedrock/minimax.minimax-m2.1.toml new file mode 100644 index 00000000000..c7cce552d5d --- /dev/null +++ b/providers/tempr/models/aws-bedrock/minimax.minimax-m2.1.toml @@ -0,0 +1,12 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# No reasoning control for this model on this host: it reasons at its own depth, +# and the x-tempr-reasoning-effort response header reports "default". +base_model = "minimax/MiniMax-M2.1" +interleaved = true +structured_output = true + +reasoning_options = [] + +[cost] +input = 0.30 +output = 1.20 diff --git a/providers/tempr/models/aws-bedrock/minimax.minimax-m2.5.toml b/providers/tempr/models/aws-bedrock/minimax.minimax-m2.5.toml new file mode 100644 index 00000000000..0dcc0624b95 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/minimax.minimax-m2.5.toml @@ -0,0 +1,17 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# No reasoning control for this model on this host: it reasons at its own depth, +# and the x-tempr-reasoning-effort response header reports "default". +base_model = "minimax/MiniMax-M2.5" +last_updated = "2026-03-18" +interleaved = true +structured_output = true + +reasoning_options = [] + +[cost] +input = 0.30 +output = 1.20 + +[limit] +context = 196_608 +output = 98_304 diff --git a/providers/tempr/models/aws-bedrock/minimax.minimax-m2.toml b/providers/tempr/models/aws-bedrock/minimax.minimax-m2.toml new file mode 100644 index 00000000000..330e67da3d1 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/minimax.minimax-m2.toml @@ -0,0 +1,16 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# No reasoning control for this model on this host: it reasons at its own depth, +# and the x-tempr-reasoning-effort response header reports "default". +base_model = "minimax/MiniMax-M2" +interleaved = true +structured_output = true + +reasoning_options = [] + +[cost] +input = 0.30 +output = 1.20 + +[limit] +context = 204_608 +output = 128_000 diff --git a/providers/tempr/models/aws-bedrock/mistral.devstral-2-123b.toml b/providers/tempr/models/aws-bedrock/mistral.devstral-2-123b.toml new file mode 100644 index 00000000000..092ed3fdc5c --- /dev/null +++ b/providers/tempr/models/aws-bedrock/mistral.devstral-2-123b.toml @@ -0,0 +1,11 @@ +base_model = "mistral/devstral-2512" +name = "Devstral 2 123B" +structured_output = true + +[cost] +input = 0.40 +output = 2.00 + +[limit] +context = 256_000 +output = 8_192 diff --git a/providers/tempr/models/aws-bedrock/mistral.magistral-small-2509.toml b/providers/tempr/models/aws-bedrock/mistral.magistral-small-2509.toml new file mode 100644 index 00000000000..7ac80c408d7 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/mistral.magistral-small-2509.toml @@ -0,0 +1,15 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# No reasoning control for this model on this host: it reasons at its own depth, +# and the x-tempr-reasoning-effort response header reports "default". +base_model = "mistral/magistral-small-2509" +structured_output = true + +reasoning_options = [] + +[cost] +input = 0.50 +output = 1.50 + +[limit] +context = 128_000 +output = 40_000 diff --git a/providers/tempr/models/aws-bedrock/mistral.ministral-3-14b-instruct.toml b/providers/tempr/models/aws-bedrock/mistral.ministral-3-14b-instruct.toml new file mode 100644 index 00000000000..a16452eb898 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/mistral.ministral-3-14b-instruct.toml @@ -0,0 +1,10 @@ +base_model = "mistral/ministral-3-14b-instruct-2512" +name = "Ministral 14B 3.0" + +[cost] +input = 0.20 +output = 0.20 + +[limit] +context = 128_000 +output = 4_096 diff --git a/providers/tempr/models/aws-bedrock/mistral.ministral-3-3b-instruct.toml b/providers/tempr/models/aws-bedrock/mistral.ministral-3-3b-instruct.toml new file mode 100644 index 00000000000..0fe95e2f69c --- /dev/null +++ b/providers/tempr/models/aws-bedrock/mistral.ministral-3-3b-instruct.toml @@ -0,0 +1,9 @@ +base_model = "mistral/ministral-3-3b-instruct-2512" + +[cost] +input = 0.10 +output = 0.10 + +[limit] +context = 256_000 +output = 8_192 diff --git a/providers/tempr/models/aws-bedrock/mistral.ministral-3-8b-instruct.toml b/providers/tempr/models/aws-bedrock/mistral.ministral-3-8b-instruct.toml new file mode 100644 index 00000000000..664de1bdfd8 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/mistral.ministral-3-8b-instruct.toml @@ -0,0 +1,9 @@ +base_model = "mistral/ministral-3-8b-instruct-2512" + +[cost] +input = 0.15 +output = 0.15 + +[limit] +context = 128_000 +output = 4_096 diff --git a/providers/tempr/models/aws-bedrock/mistral.mistral-large-3-675b-instruct.toml b/providers/tempr/models/aws-bedrock/mistral.mistral-large-3-675b-instruct.toml new file mode 100644 index 00000000000..e865b27ddf9 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/mistral.mistral-large-3-675b-instruct.toml @@ -0,0 +1,10 @@ +base_model = "mistral/mistral-large-2512" +structured_output = true + +[cost] +input = 0.50 +output = 1.50 + +[limit] +context = 256_000 +output = 8_192 diff --git a/providers/tempr/models/aws-bedrock/mistral.pixtral-large-2502-v1:0.toml b/providers/tempr/models/aws-bedrock/mistral.pixtral-large-2502-v1:0.toml new file mode 100644 index 00000000000..20a2020422a --- /dev/null +++ b/providers/tempr/models/aws-bedrock/mistral.pixtral-large-2502-v1:0.toml @@ -0,0 +1,5 @@ +base_model = "mistral/pixtral-large-2502" + +[cost] +input = 2.00 +output = 6.00 diff --git a/providers/tempr/models/aws-bedrock/mistral.voxtral-mini-3b-2507.toml b/providers/tempr/models/aws-bedrock/mistral.voxtral-mini-3b-2507.toml new file mode 100644 index 00000000000..374da0a8de6 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/mistral.voxtral-mini-3b-2507.toml @@ -0,0 +1,9 @@ +base_model = "mistral/voxtral-mini-3b-2507" +structured_output = true + +[cost] +input = 0.04 +output = 0.04 + +[limit] +output = 4_096 diff --git a/providers/tempr/models/aws-bedrock/mistral.voxtral-small-24b-2507.toml b/providers/tempr/models/aws-bedrock/mistral.voxtral-small-24b-2507.toml new file mode 100644 index 00000000000..ebb529e3e06 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/mistral.voxtral-small-24b-2507.toml @@ -0,0 +1,9 @@ +base_model = "mistral/voxtral-small-24b-2507" +structured_output = true + +[cost] +input = 0.10 +output = 0.30 + +[limit] +output = 8_192 diff --git a/providers/tempr/models/aws-bedrock/moonshot.kimi-k2-thinking.toml b/providers/tempr/models/aws-bedrock/moonshot.kimi-k2-thinking.toml new file mode 100644 index 00000000000..c8bfae50dc6 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/moonshot.kimi-k2-thinking.toml @@ -0,0 +1,17 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# No reasoning control for this model on this host: it reasons at its own depth, +# and the x-tempr-reasoning-effort response header reports "default". +base_model = "moonshotai/kimi-k2-thinking" +last_updated = "2025-12-02" +structured_output = true +interleaved = true + +reasoning_options = [] + +[cost] +input = 0.6 +output = 2.5 + +[limit] +context = 262_143 +output = 16_000 diff --git a/providers/tempr/models/aws-bedrock/moonshotai.kimi-k2.5.toml b/providers/tempr/models/aws-bedrock/moonshotai.kimi-k2.5.toml new file mode 100644 index 00000000000..5f7ebf3835c --- /dev/null +++ b/providers/tempr/models/aws-bedrock/moonshotai.kimi-k2.5.toml @@ -0,0 +1,24 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# /v1/messages: thinking.type = enabled|disabled +# /v1/responses: reasoning.effort = "none" turns it off, any other level turns it on +base_model = "moonshotai/kimi-k2.5" +release_date = "2026-01-27" +last_updated = "2026-02-06" +temperature = true +interleaved = true + +reasoning_options = [ + { type = "toggle" }, +] + +[cost] +input = 0.6 +output = 3 + +[limit] +context = 262_143 +output = 16_384 + +[modalities] +input = ["text", "image"] diff --git a/providers/tempr/models/aws-bedrock/nvidia.nemotron-nano-12b-v2.toml b/providers/tempr/models/aws-bedrock/nvidia.nemotron-nano-12b-v2.toml new file mode 100644 index 00000000000..dd0a7176f0d --- /dev/null +++ b/providers/tempr/models/aws-bedrock/nvidia.nemotron-nano-12b-v2.toml @@ -0,0 +1,14 @@ +name = "NVIDIA Nemotron Nano 12B v2 VL BF16" +base_model = "nvidia/nemotron-nano-12b-v2-vl" +reasoning = false +structured_output = true + +[cost] +input = 0.20 +output = 0.60 + +[limit] +output = 8_192 + +[modalities] +input = ["text", "image"] diff --git a/providers/tempr/models/aws-bedrock/nvidia.nemotron-nano-3-30b.toml b/providers/tempr/models/aws-bedrock/nvidia.nemotron-nano-3-30b.toml new file mode 100644 index 00000000000..3785eeb95df --- /dev/null +++ b/providers/tempr/models/aws-bedrock/nvidia.nemotron-nano-3-30b.toml @@ -0,0 +1,16 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# No reasoning control for this model on this host: it reasons at its own depth, +# and the x-tempr-reasoning-effort response header reports "default". +name = "NVIDIA Nemotron Nano 3 30B" +base_model = "nvidia/nemotron-3-nano-30b-a3b" +last_updated = "2025-12-23" +structured_output = true + +reasoning_options = [] + +[cost] +input = 0.06 +output = 0.24 + +[limit] +output = 8_192 diff --git a/providers/tempr/models/aws-bedrock/nvidia.nemotron-nano-9b-v2.toml b/providers/tempr/models/aws-bedrock/nvidia.nemotron-nano-9b-v2.toml new file mode 100644 index 00000000000..57eabf5bf4d --- /dev/null +++ b/providers/tempr/models/aws-bedrock/nvidia.nemotron-nano-9b-v2.toml @@ -0,0 +1,11 @@ +name = "NVIDIA Nemotron Nano 9B v2" +base_model = "nvidia/nemotron-nano-9b-v2" +reasoning = false +structured_output = true + +[cost] +input = 0.06 +output = 0.23 + +[limit] +output = 8_192 diff --git a/providers/tempr/models/aws-bedrock/nvidia.nemotron-super-3-120b.toml b/providers/tempr/models/aws-bedrock/nvidia.nemotron-super-3-120b.toml new file mode 100644 index 00000000000..e788110d750 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/nvidia.nemotron-super-3-120b.toml @@ -0,0 +1,15 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# No reasoning control for this model on this host: it reasons at its own depth, +# and the x-tempr-reasoning-effort response header reports "default". +name = "NVIDIA Nemotron 3 Super 120B A12B" +base_model = "nvidia/nemotron-3-super-120b-a12b" +structured_output = true + +reasoning_options = [] + +[cost] +input = 0.15 +output = 0.65 + +[limit] +output = 131_072 diff --git a/providers/tempr/models/aws-bedrock/openai.gpt-5.4.toml b/providers/tempr/models/aws-bedrock/openai.gpt-5.4.toml new file mode 100644 index 00000000000..dd26652d29f --- /dev/null +++ b/providers/tempr/models/aws-bedrock/openai.gpt-5.4.toml @@ -0,0 +1,20 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|low|medium|high|xhigh (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh +base_model = "openai/gpt-5.4" +base_model_omit = ["cost.tiers", "cost.context_over_200k", "experimental.modes.fast", "limit.input"] +last_updated = "2026-06-01" + +reasoning_options = [ + { type = "effort", values = ["none", "low", "medium", "high", "xhigh"] }, +] + +[cost] +input = 2.75 +output = 16.50 +cache_read = 0.275 + +[limit] +context = 272_000 diff --git a/providers/tempr/models/aws-bedrock/openai.gpt-5.5.toml b/providers/tempr/models/aws-bedrock/openai.gpt-5.5.toml new file mode 100644 index 00000000000..6e4a93ebd7a --- /dev/null +++ b/providers/tempr/models/aws-bedrock/openai.gpt-5.5.toml @@ -0,0 +1,20 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|low|medium|high|xhigh (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh +base_model = "openai/gpt-5.5" +base_model_omit = ["cost.tiers", "cost.context_over_200k", "experimental.modes.fast", "limit.input"] +last_updated = "2026-06-01" + +reasoning_options = [ + { type = "effort", values = ["none", "low", "medium", "high", "xhigh"] }, +] + +[cost] +input = 5.50 +output = 33.00 +cache_read = 0.55 + +[limit] +context = 272_000 diff --git a/providers/tempr/models/aws-bedrock/openai.gpt-5.6-luna.toml b/providers/tempr/models/aws-bedrock/openai.gpt-5.6-luna.toml new file mode 100644 index 00000000000..925fbb1d045 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/openai.gpt-5.6-luna.toml @@ -0,0 +1,24 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "openai/gpt-5.6-luna" +base_model_omit = ["cost.context_over_200k", "experimental.modes.fast"] + +reasoning_options = [ + { type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 0.22 +output = 1.32 +cache_read = 0.022 +cache_write = 0.275 + +[[cost.tiers]] +tier = { size = 272_000 } +input = 0.44 +output = 1.98 +cache_read = 0.044 +cache_write = 0.55 diff --git a/providers/tempr/models/aws-bedrock/openai.gpt-5.6-sol.toml b/providers/tempr/models/aws-bedrock/openai.gpt-5.6-sol.toml new file mode 100644 index 00000000000..4e188622068 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/openai.gpt-5.6-sol.toml @@ -0,0 +1,24 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "openai/gpt-5.6-sol" +base_model_omit = ["cost.context_over_200k", "experimental.modes.fast"] + +reasoning_options = [ + { type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 4.40 +output = 22.00 +cache_read = 0.44 +cache_write = 5.50 + +[[cost.tiers]] +tier = { size = 272_000 } +input = 8.80 +output = 33.00 +cache_read = 0.88 +cache_write = 11.00 diff --git a/providers/tempr/models/aws-bedrock/openai.gpt-5.6-terra.toml b/providers/tempr/models/aws-bedrock/openai.gpt-5.6-terra.toml new file mode 100644 index 00000000000..344cfbd3fad --- /dev/null +++ b/providers/tempr/models/aws-bedrock/openai.gpt-5.6-terra.toml @@ -0,0 +1,24 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "openai/gpt-5.6-terra" +base_model_omit = ["cost.context_over_200k", "experimental.modes.fast"] + +reasoning_options = [ + { type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 2.20 +output = 13.20 +cache_read = 0.22 +cache_write = 2.75 + +[[cost.tiers]] +tier = { size = 272_000 } +input = 4.40 +output = 19.80 +cache_read = 0.44 +cache_write = 5.50 diff --git a/providers/tempr/models/aws-bedrock/openai.gpt-6-astra.toml b/providers/tempr/models/aws-bedrock/openai.gpt-6-astra.toml new file mode 100644 index 00000000000..cf9d9251b4b --- /dev/null +++ b/providers/tempr/models/aws-bedrock/openai.gpt-6-astra.toml @@ -0,0 +1,23 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "openai/gpt-6-astra" +base_model_omit = ["cost.context_over_200k", "experimental.modes.fast"] + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 11.00 +output = 55.00 +cache_read = 1.10 +cache_write = 13.75 + +[[cost.tiers]] +tier = { size = 272_000 } +input = 22.00 +output = 82.50 +cache_read = 2.20 +cache_write = 27.50 diff --git a/providers/tempr/models/aws-bedrock/openai.gpt-oss-120b-1:0.toml b/providers/tempr/models/aws-bedrock/openai.gpt-oss-120b-1:0.toml new file mode 100644 index 00000000000..92a8f6c692a --- /dev/null +++ b/providers/tempr/models/aws-bedrock/openai.gpt-oss-120b-1:0.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high +# /v1/responses: reasoning.effort = low|medium|high +base_model = "openai/gpt-oss-120b" +name = "gpt-oss-120b" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high"] }, +] + +[cost] +input = 0.15 +output = 0.60 + +[limit] +context = 128_000 +output = 16_384 diff --git a/providers/tempr/models/aws-bedrock/openai.gpt-oss-120b.toml b/providers/tempr/models/aws-bedrock/openai.gpt-oss-120b.toml new file mode 100644 index 00000000000..92a8f6c692a --- /dev/null +++ b/providers/tempr/models/aws-bedrock/openai.gpt-oss-120b.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high +# /v1/responses: reasoning.effort = low|medium|high +base_model = "openai/gpt-oss-120b" +name = "gpt-oss-120b" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high"] }, +] + +[cost] +input = 0.15 +output = 0.60 + +[limit] +context = 128_000 +output = 16_384 diff --git a/providers/tempr/models/aws-bedrock/openai.gpt-oss-20b-1:0.toml b/providers/tempr/models/aws-bedrock/openai.gpt-oss-20b-1:0.toml new file mode 100644 index 00000000000..c57008c6bcd --- /dev/null +++ b/providers/tempr/models/aws-bedrock/openai.gpt-oss-20b-1:0.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high +# /v1/responses: reasoning.effort = low|medium|high +base_model = "openai/gpt-oss-20b" +name = "gpt-oss-20b" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high"] }, +] + +[cost] +input = 0.07 +output = 0.30 + +[limit] +context = 128_000 +output = 16_384 diff --git a/providers/tempr/models/aws-bedrock/openai.gpt-oss-20b.toml b/providers/tempr/models/aws-bedrock/openai.gpt-oss-20b.toml new file mode 100644 index 00000000000..c57008c6bcd --- /dev/null +++ b/providers/tempr/models/aws-bedrock/openai.gpt-oss-20b.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high +# /v1/responses: reasoning.effort = low|medium|high +base_model = "openai/gpt-oss-20b" +name = "gpt-oss-20b" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high"] }, +] + +[cost] +input = 0.07 +output = 0.30 + +[limit] +context = 128_000 +output = 16_384 diff --git a/providers/tempr/models/aws-bedrock/openai.gpt-oss-safeguard-120b.toml b/providers/tempr/models/aws-bedrock/openai.gpt-oss-safeguard-120b.toml new file mode 100644 index 00000000000..b3874db71c2 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/openai.gpt-oss-safeguard-120b.toml @@ -0,0 +1,17 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high +# /v1/responses: reasoning.effort = low|medium|high +base_model = "openai/gpt-oss-safeguard-120b" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high"] }, +] + +[cost] +input = 0.15 +output = 0.60 + +[limit] +context = 128_000 +output = 16_384 diff --git a/providers/tempr/models/aws-bedrock/openai.gpt-oss-safeguard-20b.toml b/providers/tempr/models/aws-bedrock/openai.gpt-oss-safeguard-20b.toml new file mode 100644 index 00000000000..fcd55b51d70 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/openai.gpt-oss-safeguard-20b.toml @@ -0,0 +1,17 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high +# /v1/responses: reasoning.effort = low|medium|high +base_model = "openai/gpt-oss-safeguard-20b" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high"] }, +] + +[cost] +input = 0.07 +output = 0.20 + +[limit] +context = 128_000 +output = 16_384 diff --git a/providers/tempr/models/aws-bedrock/qwen.qwen3-235b-a22b-2507-v1:0.toml b/providers/tempr/models/aws-bedrock/qwen.qwen3-235b-a22b-2507-v1:0.toml new file mode 100644 index 00000000000..e76f43f0c01 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/qwen.qwen3-235b-a22b-2507-v1:0.toml @@ -0,0 +1,10 @@ +base_model = "alibaba/qwen3-235b-a22b-instruct-2507" +last_updated = "2025-09-18" +structured_output = true + +[cost] +input = 0.22 +output = 0.88 + +[limit] +output = 131_072 diff --git a/providers/tempr/models/aws-bedrock/qwen.qwen3-32b-v1:0.toml b/providers/tempr/models/aws-bedrock/qwen.qwen3-32b-v1:0.toml new file mode 100644 index 00000000000..b4fa28306c8 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/qwen.qwen3-32b-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# /v1/messages: thinking.type = enabled|disabled +# /v1/responses: reasoning.effort = "none" turns it off, any other level turns it on +base_model = "alibaba/qwen3-32b" +last_updated = "2025-09-18" +interleaved = true +structured_output = true + +reasoning_options = [ + { type = "toggle" }, +] + +[cost] +input = 0.15 +output = 0.6 + +[limit] +context = 32_768 diff --git a/providers/tempr/models/aws-bedrock/qwen.qwen3-coder-30b-a3b-v1:0.toml b/providers/tempr/models/aws-bedrock/qwen.qwen3-coder-30b-a3b-v1:0.toml new file mode 100644 index 00000000000..fb5daf56630 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/qwen.qwen3-coder-30b-a3b-v1:0.toml @@ -0,0 +1,11 @@ +base_model = "alibaba/qwen3-coder-30b-a3b-instruct" +release_date = "2025-07-31" +last_updated = "2025-09-18" +structured_output = true + +[cost] +input = 0.15 +output = 0.6 + +[limit] +output = 131_072 diff --git a/providers/tempr/models/aws-bedrock/qwen.qwen3-coder-480b-a35b-v1:0.toml b/providers/tempr/models/aws-bedrock/qwen.qwen3-coder-480b-a35b-v1:0.toml new file mode 100644 index 00000000000..8a490f0638f --- /dev/null +++ b/providers/tempr/models/aws-bedrock/qwen.qwen3-coder-480b-a35b-v1:0.toml @@ -0,0 +1,11 @@ +base_model = "alibaba/qwen3-coder-480b-a35b-instruct" +release_date = "2025-07-23" +last_updated = "2025-09-18" +structured_output = true + +[cost] +input = 0.45 +output = 1.8 + +[limit] +context = 131_072 diff --git a/providers/tempr/models/aws-bedrock/qwen.qwen3-coder-next.toml b/providers/tempr/models/aws-bedrock/qwen.qwen3-coder-next.toml new file mode 100644 index 00000000000..1213e7014cc --- /dev/null +++ b/providers/tempr/models/aws-bedrock/qwen.qwen3-coder-next.toml @@ -0,0 +1,6 @@ +base_model = "alibaba/qwen3-coder-next" +last_updated = "2026-02-06" + +[cost] +input = 0.50 +output = 1.20 diff --git a/providers/tempr/models/aws-bedrock/qwen.qwen3-next-80b-a3b.toml b/providers/tempr/models/aws-bedrock/qwen.qwen3-next-80b-a3b.toml new file mode 100644 index 00000000000..6b2a8f25637 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/qwen.qwen3-next-80b-a3b.toml @@ -0,0 +1,12 @@ +base_model = "alibaba/qwen3-next-80b-a3b-instruct" +release_date = "2025-09-11" +last_updated = "2025-11-25" +structured_output = true + +[cost] +input = 0.15 +output = 1.20 + +[limit] +context = 262_144 +output = 262_000 diff --git a/providers/tempr/models/aws-bedrock/qwen.qwen3-vl-235b-a22b.toml b/providers/tempr/models/aws-bedrock/qwen.qwen3-vl-235b-a22b.toml new file mode 100644 index 00000000000..3da3451ca7c --- /dev/null +++ b/providers/tempr/models/aws-bedrock/qwen.qwen3-vl-235b-a22b.toml @@ -0,0 +1,10 @@ +base_model = "alibaba/qwen3-vl-235b-a22b-instruct" +last_updated = "2025-11-25" + +[cost] +input = 0.53 +output = 2.66 + +[limit] +context = 262_144 +output = 262_000 diff --git a/providers/tempr/models/aws-bedrock/us-gov.openai.gpt-oss-120b-1:0.toml b/providers/tempr/models/aws-bedrock/us-gov.openai.gpt-oss-120b-1:0.toml new file mode 100644 index 00000000000..9493cd8038e --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us-gov.openai.gpt-oss-120b-1:0.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high +# /v1/responses: reasoning.effort = low|medium|high +base_model = "openai/gpt-oss-120b" +name = "gpt-oss-120b (GovCloud)" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high"] }, +] + +[cost] +input = 0.18 +output = 0.72 + +[limit] +context = 128_000 +output = 16_384 diff --git a/providers/tempr/models/aws-bedrock/us-gov.openai.gpt-oss-20b-1:0.toml b/providers/tempr/models/aws-bedrock/us-gov.openai.gpt-oss-20b-1:0.toml new file mode 100644 index 00000000000..4cd90a8e105 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us-gov.openai.gpt-oss-20b-1:0.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high +# /v1/responses: reasoning.effort = low|medium|high +base_model = "openai/gpt-oss-20b" +name = "gpt-oss-20b (GovCloud)" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high"] }, +] + +[cost] +input = 0.084 +output = 0.36 + +[limit] +context = 128_000 +output = 16_384 diff --git a/providers/tempr/models/aws-bedrock/us.amazon.nova-2-lite-v1:0.toml b/providers/tempr/models/aws-bedrock/us.amazon.nova-2-lite-v1:0.toml new file mode 100644 index 00000000000..825cce93baf --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.amazon.nova-2-lite-v1:0.toml @@ -0,0 +1,21 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high +# /v1/responses: reasoning.effort = none|low|medium|high +base_model = "amazon/nova-2-lite" +name = "Nova 2 Lite (US)" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high"] }, +] + +[cost] +input = 0.33 +output = 2.75 +cache_read = 0.0825 +cache_write = 0.33 + +[limit] +output = 65_535 diff --git a/providers/tempr/models/aws-bedrock/us.amazon.nova-lite-v1:0.toml b/providers/tempr/models/aws-bedrock/us.amazon.nova-lite-v1:0.toml new file mode 100644 index 00000000000..2f4b363643c --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.amazon.nova-lite-v1:0.toml @@ -0,0 +1,8 @@ +base_model = "amazon/nova-lite" +name = "Nova Lite (US)" + +[cost] +input = 0.06 +output = 0.24 +cache_read = 0.015 +cache_write = 0.06 diff --git a/providers/tempr/models/aws-bedrock/us.amazon.nova-micro-v1:0.toml b/providers/tempr/models/aws-bedrock/us.amazon.nova-micro-v1:0.toml new file mode 100644 index 00000000000..417567d8940 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.amazon.nova-micro-v1:0.toml @@ -0,0 +1,8 @@ +base_model = "amazon/nova-micro" +name = "Nova Micro (US)" + +[cost] +input = 0.035 +output = 0.14 +cache_read = 0.00875 +cache_write = 0.035 diff --git a/providers/tempr/models/aws-bedrock/us.amazon.nova-premier-v1:0.toml b/providers/tempr/models/aws-bedrock/us.amazon.nova-premier-v1:0.toml new file mode 100644 index 00000000000..ea90d544b63 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.amazon.nova-premier-v1:0.toml @@ -0,0 +1,10 @@ +base_model = "amazon/nova-premier" +name = "Nova Premier (US)" +status = "deprecated" +structured_output = false + +[cost] +input = 2.50 +output = 12.50 +cache_read = 0.625 +cache_write = 2.50 diff --git a/providers/tempr/models/aws-bedrock/us.amazon.nova-pro-v1:0.toml b/providers/tempr/models/aws-bedrock/us.amazon.nova-pro-v1:0.toml new file mode 100644 index 00000000000..3f3c96e8bb8 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.amazon.nova-pro-v1:0.toml @@ -0,0 +1,8 @@ +base_model = "amazon/nova-pro" +name = "Nova Pro (US)" + +[cost] +input = 0.80 +output = 3.20 +cache_read = 0.20 +cache_write = 0.80 diff --git a/providers/tempr/models/aws-bedrock/us.anthropic.claude-fable-5-1.toml b/providers/tempr/models/aws-bedrock/us.anthropic.claude-fable-5-1.toml new file mode 100644 index 00000000000..7df8ca37fa3 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.anthropic.claude-fable-5-1.toml @@ -0,0 +1,16 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "anthropic/claude-fable-5-1" +name = "Claude Fable 5.1 (US)" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 11 +output = 55 +cache_read = 0.275 +cache_write = 13.75 diff --git a/providers/tempr/models/aws-bedrock/us.anthropic.claude-fable-5.toml b/providers/tempr/models/aws-bedrock/us.anthropic.claude-fable-5.toml new file mode 100644 index 00000000000..d820d2c498f --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.anthropic.claude-fable-5.toml @@ -0,0 +1,16 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "anthropic/claude-fable-5" +name = "Claude Fable 5 (US)" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 11 +output = 55 +cache_read = 1.1 +cache_write = 13.75 diff --git a/providers/tempr/models/aws-bedrock/us.anthropic.claude-haiku-4-5-20251001-v1:0.toml b/providers/tempr/models/aws-bedrock/us.anthropic.claude-haiku-4-5-20251001-v1:0.toml new file mode 100644 index 00000000000..78d3f4145ff --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.anthropic.claude-haiku-4-5-20251001-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-haiku-4-5-20251001" +name = "Claude Haiku 4.5 (US)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 1.10 +output = 5.50 +cache_read = 0.11 +cache_write = 1.375 diff --git a/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-4-1-20250805-v1:0.toml b/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-4-1-20250805-v1:0.toml new file mode 100644 index 00000000000..d04ab364308 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-4-1-20250805-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-opus-4-1-20250805" +status = "deprecated" +name = "Claude Opus 4.1 (US)" + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 15 +output = 75 +cache_read = 1.5 +cache_write = 18.75 diff --git a/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-4-5-20251101-v1:0.toml b/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-4-5-20251101-v1:0.toml new file mode 100644 index 00000000000..a5d4a3afb5d --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-4-5-20251101-v1:0.toml @@ -0,0 +1,21 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high (alias: top-level reasoning_effort) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high +base_model = "anthropic/claude-opus-4-5-20251101" +name = "Claude Opus 4.5 (US)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high"] }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 5.50 +output = 27.50 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-4-6-v1.toml b/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-4-6-v1.toml new file mode 100644 index 00000000000..ea252810e5a --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-4-6-v1.toml @@ -0,0 +1,21 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|max (alias: top-level reasoning_effort) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|max; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high|max +base_model = "anthropic/claude-opus-4-6" +name = "Claude Opus 4.6 (US)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "max"] }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 5.50 +output = 27.50 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-4-7.toml b/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-4-7.toml new file mode 100644 index 00000000000..f6f6583a846 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-4-7.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-4-7" +name = "Claude Opus 4.7 (US)" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5.50 +output = 27.50 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-4-8.toml b/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-4-8.toml new file mode 100644 index 00000000000..4fcaa79e744 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-4-8.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-4-8" +name = "Claude Opus 4.8 (US)" + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5.50 +output = 27.50 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-5.toml b/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-5.toml new file mode 100644 index 00000000000..061fb3314dd --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.anthropic.claude-opus-5.toml @@ -0,0 +1,16 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "anthropic/claude-opus-5" +name = "Claude Opus 5 (US)" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 5.50 +output = 27.50 +cache_read = 0.55 +cache_write = 6.875 diff --git a/providers/tempr/models/aws-bedrock/us.anthropic.claude-sonnet-4-20250514-v1:0.toml b/providers/tempr/models/aws-bedrock/us.anthropic.claude-sonnet-4-20250514-v1:0.toml new file mode 100644 index 00000000000..c88c17cc185 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.anthropic.claude-sonnet-4-20250514-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-sonnet-4-20250514" +name = "Claude Sonnet 4 (US)" +status = "deprecated" + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3 +output = 15 +cache_read = 0.3 +cache_write = 3.75 diff --git a/providers/tempr/models/aws-bedrock/us.anthropic.claude-sonnet-4-5-20250929-v1:0.toml b/providers/tempr/models/aws-bedrock/us.anthropic.claude-sonnet-4-5-20250929-v1:0.toml new file mode 100644 index 00000000000..f887e0dd7b9 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.anthropic.claude-sonnet-4-5-20250929-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high (sized to a thinking budget) +base_model = "anthropic/claude-sonnet-4-5-20250929" +name = "Claude Sonnet 4.5 (US)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3.30 +output = 16.50 +cache_read = 0.33 +cache_write = 4.125 diff --git a/providers/tempr/models/aws-bedrock/us.anthropic.claude-sonnet-4-6.toml b/providers/tempr/models/aws-bedrock/us.anthropic.claude-sonnet-4-6.toml new file mode 100644 index 00000000000..2e7409ddf8f --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.anthropic.claude-sonnet-4-6.toml @@ -0,0 +1,24 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|max (alias: top-level reasoning_effort) +# Budget: reasoning.max_tokens (integer reasoning tokens) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|max; thinking.budget_tokens +# /v1/responses: reasoning.effort = none|low|medium|high|max +base_model = "anthropic/claude-sonnet-4-6" +name = "Claude Sonnet 4.6 (US)" +structured_output = true + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "max"] }, + { type = "budget_tokens", min = 1024 }, +] + +[cost] +input = 3.30 +output = 16.50 +cache_read = 0.33 +cache_write = 4.125 + +[limit] +output = 128_000 diff --git a/providers/tempr/models/aws-bedrock/us.anthropic.claude-sonnet-5.toml b/providers/tempr/models/aws-bedrock/us.anthropic.claude-sonnet-5.toml new file mode 100644 index 00000000000..6c3026789c1 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.anthropic.claude-sonnet-5.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "anthropic/claude-sonnet-5" +name = "Claude Sonnet 5 (US)" +structured_output = false + +reasoning_options = [ + { type = "toggle" }, + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 2.20 +output = 11 +cache_read = 0.22 +cache_write = 2.75 diff --git a/providers/tempr/models/aws-bedrock/us.deepseek.r1-v1:0.toml b/providers/tempr/models/aws-bedrock/us.deepseek.r1-v1:0.toml new file mode 100644 index 00000000000..fff870c6de1 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.deepseek.r1-v1:0.toml @@ -0,0 +1,12 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# No reasoning control for this model on this host: it reasons at its own depth, +# and the x-tempr-reasoning-effort response header reports "default". +base_model = "deepseek/deepseek-r1" +name = "DeepSeek-R1 (US)" +tool_call = false + +reasoning_options = [] + +[cost] +input = 1.35 +output = 5.4 diff --git a/providers/tempr/models/aws-bedrock/us.meta.llama3-1-70b-instruct-v1:0.toml b/providers/tempr/models/aws-bedrock/us.meta.llama3-1-70b-instruct-v1:0.toml new file mode 100644 index 00000000000..3f7a0aa755d --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.meta.llama3-1-70b-instruct-v1:0.toml @@ -0,0 +1,6 @@ +base_model = "meta/llama-3.1-70b-instruct" +name = "Llama 3.1 70B Instruct (US)" + +[cost] +input = 0.72 +output = 0.72 diff --git a/providers/tempr/models/aws-bedrock/us.meta.llama3-1-8b-instruct-v1:0.toml b/providers/tempr/models/aws-bedrock/us.meta.llama3-1-8b-instruct-v1:0.toml new file mode 100644 index 00000000000..c1cc6e98bdb --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.meta.llama3-1-8b-instruct-v1:0.toml @@ -0,0 +1,6 @@ +base_model = "meta/llama-3.1-8b-instruct" +name = "Llama 3.1 8B Instruct (US)" + +[cost] +input = 0.22 +output = 0.22 diff --git a/providers/tempr/models/aws-bedrock/us.meta.llama3-3-70b-instruct-v1:0.toml b/providers/tempr/models/aws-bedrock/us.meta.llama3-3-70b-instruct-v1:0.toml new file mode 100644 index 00000000000..57d57ad715c --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.meta.llama3-3-70b-instruct-v1:0.toml @@ -0,0 +1,6 @@ +base_model = "meta/llama-3.3-70b-instruct" +name = "Llama 3.3 70B Instruct (US)" + +[cost] +input = 0.72 +output = 0.72 diff --git a/providers/tempr/models/aws-bedrock/us.meta.llama4-maverick-17b-instruct-v1:0.toml b/providers/tempr/models/aws-bedrock/us.meta.llama4-maverick-17b-instruct-v1:0.toml new file mode 100644 index 00000000000..3e7d59c4c70 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.meta.llama4-maverick-17b-instruct-v1:0.toml @@ -0,0 +1,9 @@ +base_model = "meta/llama-4-maverick-17b-instruct" +name = "Llama 4 Maverick 17B Instruct (US)" + +[cost] +input = 0.24 +output = 0.97 + +[limit] +output = 8_192 diff --git a/providers/tempr/models/aws-bedrock/us.meta.llama4-scout-17b-instruct-v1:0.toml b/providers/tempr/models/aws-bedrock/us.meta.llama4-scout-17b-instruct-v1:0.toml new file mode 100644 index 00000000000..0aa1ef522cf --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.meta.llama4-scout-17b-instruct-v1:0.toml @@ -0,0 +1,9 @@ +base_model = "meta/llama-4-scout-17b-instruct" +name = "Llama 4 Scout 17B Instruct (US)" + +[cost] +input = 0.17 +output = 0.66 + +[limit] +output = 8_192 diff --git a/providers/tempr/models/aws-bedrock/us.mistral.pixtral-large-2502-v1:0.toml b/providers/tempr/models/aws-bedrock/us.mistral.pixtral-large-2502-v1:0.toml new file mode 100644 index 00000000000..f73101e09ba --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.mistral.pixtral-large-2502-v1:0.toml @@ -0,0 +1,6 @@ +base_model = "mistral/pixtral-large-2502" +name = "Pixtral Large (25.02) (US)" + +[cost] +input = 2.00 +output = 6.00 diff --git a/providers/tempr/models/aws-bedrock/us.openai.gpt-5.6-luna.toml b/providers/tempr/models/aws-bedrock/us.openai.gpt-5.6-luna.toml new file mode 100644 index 00000000000..724405ca627 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.openai.gpt-5.6-luna.toml @@ -0,0 +1,28 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "openai/gpt-5.6-luna" +base_model_omit = ["cost.context_over_200k", "experimental.modes.fast"] +name = "GPT-5.6 Luna (US)" + +reasoning_options = [ + { type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 0.22 +output = 1.32 +cache_read = 0.022 +cache_write = 0.275 + +[[cost.tiers]] +tier = { size = 272_000 } +input = 0.44 +output = 1.98 +cache_read = 0.044 +cache_write = 0.55 + +[modalities] +input = ["text", "image"] diff --git a/providers/tempr/models/aws-bedrock/us.openai.gpt-5.6-sol.toml b/providers/tempr/models/aws-bedrock/us.openai.gpt-5.6-sol.toml new file mode 100644 index 00000000000..375714cecbb --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.openai.gpt-5.6-sol.toml @@ -0,0 +1,29 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "openai/gpt-5.6-sol" +base_model_omit = ["cost.context_over_200k", "experimental.modes.fast"] +name = "GPT-5.6 Sol (US)" +structured_output = false + +reasoning_options = [ + { type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 4.40 +output = 22.00 +cache_read = 0.44 +cache_write = 5.50 + +[[cost.tiers]] +tier = { size = 272_000 } +input = 8.80 +output = 33.00 +cache_read = 0.88 +cache_write = 11.00 + +[modalities] +input = ["text", "image"] diff --git a/providers/tempr/models/aws-bedrock/us.openai.gpt-5.6-terra.toml b/providers/tempr/models/aws-bedrock/us.openai.gpt-5.6-terra.toml new file mode 100644 index 00000000000..700a06775aa --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.openai.gpt-5.6-terra.toml @@ -0,0 +1,29 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = none|low|medium|high|xhigh|max +base_model = "openai/gpt-5.6-terra" +base_model_omit = ["cost.context_over_200k", "experimental.modes.fast"] +name = "GPT-5.6 Terra (US)" +structured_output = false + +reasoning_options = [ + { type = "effort", values = ["none", "low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 2.20 +output = 13.20 +cache_read = 0.22 +cache_write = 2.75 + +[[cost.tiers]] +tier = { size = 272_000 } +input = 4.40 +output = 19.80 +cache_read = 0.44 +cache_write = 5.50 + +[modalities] +input = ["text", "image"] diff --git a/providers/tempr/models/aws-bedrock/us.openai.gpt-6-astra.toml b/providers/tempr/models/aws-bedrock/us.openai.gpt-6-astra.toml new file mode 100644 index 00000000000..96702502303 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.openai.gpt-6-astra.toml @@ -0,0 +1,27 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh|max (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh|max +# /v1/responses: reasoning.effort = low|medium|high|xhigh|max +base_model = "openai/gpt-6-astra" +base_model_omit = ["cost.context_over_200k", "experimental.modes.fast"] +name = "GPT-6 Astra (US)" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh", "max"] }, +] + +[cost] +input = 11.00 +output = 55.00 +cache_read = 1.10 +cache_write = 13.75 + +[[cost.tiers]] +tier = { size = 272_000 } +input = 22.00 +output = 82.50 +cache_read = 2.20 +cache_write = 27.50 + +[modalities] +input = ["text", "image"] diff --git a/providers/tempr/models/aws-bedrock/us.writer.palmyra-x4-v1:0.toml b/providers/tempr/models/aws-bedrock/us.writer.palmyra-x4-v1:0.toml new file mode 100644 index 00000000000..f741d15f401 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.writer.palmyra-x4-v1:0.toml @@ -0,0 +1,15 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# No reasoning control for this model on this host: it reasons at its own depth, +# and the x-tempr-reasoning-effort response header reports "default". +base_model = "writer/palmyra-x4" +name = "Palmyra X4 (US)" + +reasoning_options = [] + +[cost] +input = 2.50 +output = 10.00 + +[limit] +context = 122_880 +output = 8_192 diff --git a/providers/tempr/models/aws-bedrock/us.writer.palmyra-x5-v1:0.toml b/providers/tempr/models/aws-bedrock/us.writer.palmyra-x5-v1:0.toml new file mode 100644 index 00000000000..6bd01ecee9f --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.writer.palmyra-x5-v1:0.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# No reasoning control for this model on this host: it reasons at its own depth, +# and the x-tempr-reasoning-effort response header reports "default". +base_model = "writer/palmyra-x5" +attachment = false +name = "Palmyra X5 (US)" + +reasoning_options = [] + +[cost] +input = 0.60 +output = 6.00 + +[limit] +context = 1_040_000 +input = 1_040_000 + +[modalities] +input = ["text"] diff --git a/providers/tempr/models/aws-bedrock/us.xai.grok-4.6.toml b/providers/tempr/models/aws-bedrock/us.xai.grok-4.6.toml new file mode 100644 index 00000000000..d4b7953b89c --- /dev/null +++ b/providers/tempr/models/aws-bedrock/us.xai.grok-4.6.toml @@ -0,0 +1,17 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh +# /v1/responses: reasoning.effort = low|medium|high|xhigh +base_model = "xai/grok-4.6" +name = "Grok 4.6 (US)" +last_updated = "2026-08-18" +structured_output = false + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh"] }, +] + +[cost] +input = 2.20 +output = 6.60 +cache_read = 0.55 diff --git a/providers/tempr/models/aws-bedrock/writer.palmyra-x4-v1:0.toml b/providers/tempr/models/aws-bedrock/writer.palmyra-x4-v1:0.toml new file mode 100644 index 00000000000..cdee11a9866 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/writer.palmyra-x4-v1:0.toml @@ -0,0 +1,14 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# No reasoning control for this model on this host: it reasons at its own depth, +# and the x-tempr-reasoning-effort response header reports "default". +base_model = "writer/palmyra-x4" + +reasoning_options = [] + +[cost] +input = 2.5 +output = 10 + +[limit] +context = 122_880 +output = 8_192 diff --git a/providers/tempr/models/aws-bedrock/writer.palmyra-x5-v1:0.toml b/providers/tempr/models/aws-bedrock/writer.palmyra-x5-v1:0.toml new file mode 100644 index 00000000000..a0dbb261872 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/writer.palmyra-x5-v1:0.toml @@ -0,0 +1,18 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# No reasoning control for this model on this host: it reasons at its own depth, +# and the x-tempr-reasoning-effort response header reports "default". +base_model = "writer/palmyra-x5" +attachment = false + +reasoning_options = [] + +[cost] +input = 0.6 +output = 6 + +[limit] +context = 1_040_000 +input = 1_040_000 + +[modalities] +input = ["text"] diff --git a/providers/tempr/models/aws-bedrock/xai.grok-4.3.toml b/providers/tempr/models/aws-bedrock/xai.grok-4.3.toml new file mode 100644 index 00000000000..cbdefdae38b --- /dev/null +++ b/providers/tempr/models/aws-bedrock/xai.grok-4.3.toml @@ -0,0 +1,22 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Off: reasoning.effort = "none" (or reasoning.enabled = false) +# Effort: reasoning.effort = none|low|medium|high (alias: top-level reasoning_effort) +# /v1/messages: thinking.type = enabled|disabled; output_config.effort = low|medium|high +# /v1/responses: reasoning.effort = none|low|medium|high +base_model = "xai/grok-4.3" +last_updated = "2026-06-28" + +reasoning_options = [ + { type = "effort", values = ["none", "low", "medium", "high"] }, +] + +[cost] +input = 1.25 +output = 2.50 +cache_read = 0.20 + +[limit] +output = 131_072 + +[modalities] +input = ["text", "image"] diff --git a/providers/tempr/models/aws-bedrock/xai.grok-4.6.toml b/providers/tempr/models/aws-bedrock/xai.grok-4.6.toml new file mode 100644 index 00000000000..e12c8cd903c --- /dev/null +++ b/providers/tempr/models/aws-bedrock/xai.grok-4.6.toml @@ -0,0 +1,15 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Effort: reasoning.effort = low|medium|high|xhigh (alias: top-level reasoning_effort) +# /v1/messages: output_config.effort = low|medium|high|xhigh +# /v1/responses: reasoning.effort = low|medium|high|xhigh +base_model = "xai/grok-4.6" +last_updated = "2026-08-18" + +reasoning_options = [ + { type = "effort", values = ["low", "medium", "high", "xhigh"] }, +] + +[cost] +input = 2.20 +output = 6.60 +cache_read = 0.55 diff --git a/providers/tempr/models/aws-bedrock/zai.glm-4.7-flash.toml b/providers/tempr/models/aws-bedrock/zai.glm-4.7-flash.toml new file mode 100644 index 00000000000..b7285a969ba --- /dev/null +++ b/providers/tempr/models/aws-bedrock/zai.glm-4.7-flash.toml @@ -0,0 +1,15 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# /v1/messages: thinking.type = enabled|disabled +# /v1/responses: reasoning.effort = "none" turns it off, any other level turns it on +base_model = "zhipuai/glm-4.7-flash" +interleaved = true +structured_output = true + +reasoning_options = [ + { type = "toggle" }, +] + +[cost] +input = 0.07 +output = 0.40 diff --git a/providers/tempr/models/aws-bedrock/zai.glm-4.7.toml b/providers/tempr/models/aws-bedrock/zai.glm-4.7.toml new file mode 100644 index 00000000000..ecc758d0b53 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/zai.glm-4.7.toml @@ -0,0 +1,15 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# /v1/messages: thinking.type = enabled|disabled +# /v1/responses: reasoning.effort = "none" turns it off, any other level turns it on +base_model = "zhipuai/glm-4.7" +structured_output = true +interleaved = true + +reasoning_options = [ + { type = "toggle" }, +] + +[cost] +input = 0.60 +output = 2.20 diff --git a/providers/tempr/models/aws-bedrock/zai.glm-5.toml b/providers/tempr/models/aws-bedrock/zai.glm-5.toml new file mode 100644 index 00000000000..d4eb59b4cb3 --- /dev/null +++ b/providers/tempr/models/aws-bedrock/zai.glm-5.toml @@ -0,0 +1,19 @@ +# Tempr Gateway: https://temprhq.io/docs/gateway-chat-completions#reasoning +# Toggle: reasoning.enabled = true|false (reasoning.effort = "none" also turns it off) +# /v1/messages: thinking.type = enabled|disabled +# /v1/responses: reasoning.effort = "none" turns it off, any other level turns it on +base_model = "zhipuai/glm-5" +last_updated = "2026-03-18" +structured_output = true +interleaved = true + +reasoning_options = [ + { type = "toggle" }, +] + +[cost] +input = 1.00 +output = 3.20 + +[limit] +context = 202_752 diff --git a/providers/tempr/provider.toml b/providers/tempr/provider.toml new file mode 100644 index 00000000000..70ceee60f53 --- /dev/null +++ b/providers/tempr/provider.toml @@ -0,0 +1,5 @@ +name = "Tempr" +npm = "@ai-sdk/openai-compatible" +env = ["TEMPR_API_KEY"] +api = "https://api.temprhq.io/v1" +doc = "https://temprhq.io/docs/gateway-reference.html"