Models
The available models are whatever GET /v1/models returns right now; the catalog below lists the commonly used model IDs.
GPT6
gpt-6-astragpt-5.6-solgpt-5.6-terragpt-5.6-lunagpt-5.5gpt-5.4
Claude8
claude-opus-5claude-fable-5claude-sonnet-5claude-opus-4-8claude-opus-4-7claude-opus-4-6claude-sonnet-4-6claude-haiku-4-5
Gemini6
gemini-3.8-flashgemini-3.7-flashgemini-3.6-flashgemini-3.5-flash-litegemini-3.1-pro-previewgemini-3.1-flash-lite-preview
DeepSeek4
deepseek-flashdeepseek-v4-flash-vision-expdeepseek-v4-prodeepseek-v4-flash
GLM3
glm-5.3-flashglm-5.3glm-5.2
Grok1
grok-4.6
Models API standard
GET /v1/models returns every model you can call, and its spelling of model IDs is authoritative; see List models. IDs with a [1M] suffix are long-context variants of the same model; see 1M context.
API response schema
One entry per model: id (the ID to use in requests), object (always model), and owned_by (the vendor).
Model pricing
Prices come from GET /v1/models/pricing; see Get model pricing, or the pricing page on the web.