`mistral://`¶

Mistral AI — Mistral Large, Small, Codestral, Embed, etc. via Mistral's OpenAI-compatible chat-completions API.

Built on pkg/component/llmutil/ — the same chat/embeddings shape as openai://. Wire format, headers, and timeout semantics are inherited from there; this page documents the Mistral-specific defaults.

URI shape¶

mistral://chat?model=<id>&apiKey=<key>&temperature=<float>&maxTokens=<int>&baseURL=<url>
mistral://embeddings?model=<id>&apiKey=<key>

Parameters¶

Param	Type	Default	Producer/Consumer/Both	Description
path	enum	required	Producer	`chat` or `embeddings`.
`model`	string	`mistral-large-latest`	Producer	Mistral model id. Common chat: `mistral-large-latest`, `mistral-small-latest`, `open-mixtral-8x22b`, `codestral-latest`. Embeddings: `mistral-embed`.
`apiKey`	string	(from constructor)	Producer	Mistral API key. Sent as `Authorization: Bearer <key>`.
`baseURL`	string	`https://api.mistral.ai/v1`	Producer	Override for proxies.
`temperature`	float	`0.7`	Producer	Sampling temperature.
`maxTokens`	int	`4096`	Producer	Output cap.

Producer-only. Default timeout 120 s.

Headers¶

Same as openai:// but namespaced under mistral.* (mistral.model, mistral.id, mistral.finishReason, mistral.usage.promptTokens, mistral.usage.completionTokens, mistral.usage.totalTokens). Input header systemPrompt works the same way.

Examples¶

- to: mistral://chat?model=mistral-large-latest&apiKey=${MISTRAL_API_KEY}

Code generation via Codestral:

- to: mistral://chat?model=codestral-latest&apiKey=${MISTRAL_API_KEY}&temperature=0.2

Embeddings:

- to: mistral://embeddings?model=mistral-embed&apiKey=${MISTRAL_API_KEY}

Behavior notes¶

See openai:// for the shared behaviour — no streaming on this raw client, no tool use, single-turn, 120 s timeout. The full Mistral model list is at https://docs.mistral.ai/getting-started/models/models_overview/.

Source¶

pkg/component/mistral/mistral.go — eight-line wrapper around llmutil.NewComponent.
pkg/component/llmutil/llmutil.go — shared base.

mistral://¶