From cfee97fd5763db0e0334d25739306db307a4915a Mon Sep 17 00:00:00 2001 From: dylanvu Date: Sat, 19 Sep 2026 15:00:21 -0400 Subject: [PATCH] feat: add Pareto Inference metered Flash API --- providers/pareto/logo.svg | 5 +++++ .../pareto/models/z-ai/glm-5.3-flash.toml | 18 ++++++++++++++++++ providers/pareto/provider.toml | 6 ++++++ 3 files changed, 29 insertions(+) create mode 100644 providers/pareto/logo.svg create mode 100644 providers/pareto/models/z-ai/glm-5.3-flash.toml create mode 100644 providers/pareto/provider.toml diff --git a/providers/pareto/logo.svg b/providers/pareto/logo.svg new file mode 100644 index 00000000000..cf140760f7e --- /dev/null +++ b/providers/pareto/logo.svg @@ -0,0 +1,5 @@ + + + + + diff --git a/providers/pareto/models/z-ai/glm-5.3-flash.toml b/providers/pareto/models/z-ai/glm-5.3-flash.toml new file mode 100644 index 00000000000..0640a69a112 --- /dev/null +++ b/providers/pareto/models/z-ai/glm-5.3-flash.toml @@ -0,0 +1,18 @@ +# Pricing: https://docs.paretoinference.com/pricing (checked 2026-09-19). +# API: https://docs.paretoinference.com/api-reference +# Effort: reasoning_effort = low|high|max; no thinking/reasoning toggle. +# Context follows the 1M dashboard model entry; output is capped at 131,072. +# PDF input is not advertised on this API. +base_model = "zhipuai/glm-5.3-flash" +reasoning_options = [{ type = "effort", values = ["low", "high", "max"] }] + +[interleaved] +field = "reasoning_content" + +[cost] +input = 0.03 +output = 0.10 +cache_read = 0.006 + +[modalities] +input = ["text", "image", "video"] diff --git a/providers/pareto/provider.toml b/providers/pareto/provider.toml new file mode 100644 index 00000000000..cf3f319e0ba --- /dev/null +++ b/providers/pareto/provider.toml @@ -0,0 +1,6 @@ +# Prepaid, metered API: https://docs.paretoinference.com/pricing +name = "Pareto Inference" +npm = "@ai-sdk/openai-compatible" +env = ["PARETO_API_KEY"] +api = "https://api.paretoinference.com/v1" +doc = "https://docs.paretoinference.com/"