---
title: "Check Chat Model Health"
method: GET
path: "/health/chat-model"
tags: ["health"]
---

# Check Chat Model Health

`GET /health/chat-model`

Health check for a specific chat model.

Runs the full chat pipeline (fetch user details, memories, chat history,
build system instruction, call LLM) but skips all database writes.
Returns model status, latency breakdown, and a response preview.

## Query parameters

- `model` string, required — Model name, e.g. 'gemini-2.5-flash', 'openai:gpt-4o', 'azure:gpt-4o-mini'
- `user_id` string, nullable — User ID for fetching real user context (reads only, no writes)

## Headers

- `x-api-key` string, required

## Response `200`

Successful Response

- unknown

## Other responses

- `422` — Validation Error

---

[API](https://skmtc.net/usecreed/apis/creed-api.md) · [All operations](https://skmtc.net/usecreed/apis/creed-api/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/usecreed/creed-api/versions/523833f9e819/schema)