[Feature]: Support for Mistral as LLM provider

[JosephChataignon]

Problem or Use Case

Mistral is currently not among the providers natively supported by Hermes-Agent. It has a bigger user base than some of the already-supported providers, and its voice models are already integrated, so the LLMs shouldn't be too difficult to add.

Proposed Solution

Add Mistral to the list of existing LLM providers.
API endpoint: https://api.mistral.ai/v1/

Alternatives Considered

Currently one can use Mistral with the "custom provider" option, which is less convenient

Feature Type

Configuration option

Scope

Small (single file, < 50 lines)

Contribution

• I'd like to implement this myself and submit a PR

Debug Report (optional)

[alt-glitch]

Related: PR #7856 adds Mistral AI as a first-class provider, and PR #4485 bundles Mistral with Nebius + Scaleway. This request is covered by those open PRs.

See also: #11455 (Mistral API compat patches), #7852 (reasoning support for Mistral).

[mhorlacher]

Still waiting for this..

[rcp-family]

Still waiting for this [2]

[alainsanguinetti]

Also waiting for this [3]

[luluthehungrycat]

This is now addressed by PR #65450 — a bundled Mistral AI model provider plugin (api.mistral.ai/v1) with auto-registration, tool calling, vision, reasoning_effort gating (version-threshold based — future models covered automatically), and 48 unit tests.

Simply set MISTRAL_API_KEY and the provider is auto-discovered. No custom provider config needed.

#65450

[QuentinDEME97]

I see that the mentioned PR is closed. Any updates ?