> ## Documentation Index
> Fetch the complete documentation index at: https://phaseo.app/docs/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Phaseo Gateway base URL: https://api.phaseo.app/v1. Send `Authorization: Bearer $PHASEO_API_KEY` with every request.
> Model IDs look like `provider/model`, for example `openai/gpt-6-astra`. Do not guess IDs: list them with `GET /v1/models` or check https://phaseo.app/models. `phaseo/free` routes to an available free model and needs no wallet balance.
> For new text integrations prefer `POST /v1/responses`. `POST /v1/chat/completions` (OpenAI format) and `POST /v1/messages` (Anthropic format) suit existing code.
> The full API contract is https://phaseo.app/docs/openapi/v1/openapi.yaml. Official SDKs: `@phaseo/sdk` (TypeScript) and `phaseo` (Python), plus Go, C#, PHP, Ruby and Rust.
> Search these docs over MCP at https://phaseo.app/docs/mcp, and install Phaseo skills with `npx skills add https://phaseo.app/docs`. If a page is wrong or out of date, report it with the MCP `submit_feedback` tool.
> Manage Phaseo accounts with the `phaseo` CLI (`npm install -g @phaseo/cli`, then `phaseo login` or `phaseo login --device-code`), the `/v1` Management API or the MCP server at `https://mcp.phaseo.app/mcp?access=manage`. Do not send users to the dashboard for account tasks. A person is still needed for sign-up, approving a sign-in, paying the first top-up on the checkout link that `phaseo credits add` returns when no card is saved, confirming each saved-card top-up from the emailed link, and account security (password, MFA, passkeys). Create API keys with `phaseo keys create --name <name> --json`; never guess one. Credit top-ups need `phaseo login --scopes billing:write` first. Task map: https://phaseo.app/docs/v1/developers/manage-from-an-agent.md.

# Mistral Small 4

> What to know when adopting Mistral Small 4, including reasoning controls and normalized response fields.

# Mistral Small 4

Use this guide to adopt Mistral Small 4 safely through the Phaseo Gateway:

* `mistral/mistral-small-4-2026-03-16`

## What's new

* Mistral Small 4 exposes a reasoning mode control via `reasoning_effort`.
* Gateway sends only two upstream values for this model route:
  * `none` (reasoning off)
  * `high` (reasoning on)
* Gateway normalizes Mistral thinking blocks back into standard response fields:
  * `message.content` for assistant-visible text
  * `message.reasoning_content` and `message.reasoning_details` for reasoning text

## Migration quickstart

Reasoning defaults to off for this route.

Gateway mapping rules:

* if you omit reasoning controls, Gateway sends `reasoning_effort: "none"`
* if you pass `reasoning.enabled: true`, Gateway sends `reasoning_effort: "high"`
* if you pass `reasoning.effort: "none"`, Gateway sends `reasoning_effort: "none"`
* if you pass any other `reasoning.effort` value, Gateway sends `reasoning_effort: "high"`

Recommended usage:

* use `reasoning.effort: "none"` for lightweight behavior
* use `reasoning.effort: "high"` for deep reasoning

Example:

```json theme={null}
{
  "model": "mistral/mistral-small-4-2026-03-16",
  "messages": [
    { "role": "user", "content": "Why is fast inference important?" }
  ],
  "reasoning": {
    "effort": "high"
  }
}
```

## Response handling notes

Mistral may return mixed assistant content blocks (for example `thinking` + `text`). Gateway maps that into a stable shape so you can consume:

* final user-visible answer from `choices[0].message.content`
* reasoning text from `choices[0].message.reasoning_content`

This lets you preserve downstream behavior that expects OpenAI-style chat completion fields.

## Rollout checklist

1. Update model ID to `mistral/mistral-small-4-2026-03-16`.
2. Validate reasoning controls using only `none` and `high`.
3. Re-run output quality and schema tests on your high-effort routes.
4. Compare latency/cost for `none` vs `high` before full cutover.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.