feat: add Pareto Inference metered GLM 5.3 Flash API

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"]

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/"