[bug] Incorrect API endpoint handling in LM Studio's server.

[mirpo]

Which version of LM Studio?

LM Studio 0.3.14

Which operating system?

Mac

What is the bug?

Incorrect API endpoint handling in LM Studio's server. When attempting to use the /chat/completions endpoint (without the /v1/ prefix), the server returns a 200 OK response with an error message instead of the appropriate 404 status code.

Technical details

When making a POST request to http://localhost:1234/chat/completions, the server responds with a 200 OK status code but returns an error message:

{"error":"Unexpected endpoint or method. (POST /chat/completions)"}

The correct endpoint that works is http://localhost:1234/v1/chat/completions, following the OpenAI API format.

Screenshots

---

Logs

< HTTP/1.1 200 OK
< X-Powered-By: Express
< Content-Type: application/json; charset=utf-8
< Content-Length: 67
< ETag: W/"43-qTdCaLfEMe8wb/ShYO0yTUTunEg"
< Date: Fri, 25 Apr 2025 18:22:52 GMT
< Connection: keep-alive
< Keep-Alive: timeout=5
<
* Connection #0 to host localhost left intact
{"error":"Unexpected endpoint or method. (POST /chat/completions)"}

To Reproduce

1. Start LM Studio server on port 1234
2. Run the following curl command:

curl --verbose http://localhost:1234/chat/completions \
    -H "Content-Type: application/json" \
    -d '{
        "model": "any-model-name",
        "messages": [
            { "role": "system", "content": "Always answer in rhymes. Today is Thursday" },
            { "role": "user", "content": "What day is it today?" }
        ],
        "temperature": 0.7,
        "max_tokens": -1,
        "stream": false
    }'

3. Observe that the server returns a 200 OK status with an error message
   {"error":"Unexpected endpoint or method. (POST /chat/completions)"}

Expected behavior

The server should either:

1. Return a 404 status code for invalid endpoints, or
2. Support both /chat/completions and /v1/chat/completions as valid endpoints

Working example

The following endpoint works correctly:

curl --verbose http://localhost:1234/v1/chat/completions \
    -H "Content-Type: application/json" \
    -d '{
        "model": "any-model-name",
        "messages": [
            { "role": "system", "content": "Always answer in rhymes. Today is Thursday" },
            { "role": "user", "content": "What day is it today?" }
        ],
        "temperature": 0.7,
        "max_tokens": -1,
        "stream": false
    }'

Additional information

This issue can cause confusion for developers transitioning from other API implementations or following older documentation. The server responds with a success status code (200) despite the request failing, which violates HTTP conventions and makes error handling more difficult for client applications.

[ramaraoganji-spec]

2025-11-07 14:28:06 [ERROR]
Unexpected endpoint or method. (GET /favicon.ico). Returning 200 anyway

[amor-ai-lab]

curl http://localhost:1234/api/v1/chat/completions \ -H "Content-Type: application/json" \ -d '{ "model": "text-embedding-nomic-embed-text-v1.5", "system_prompt": "You answer only in rhymes.", "input": "What is your favorite color?" }' {"error":"Unexpected endpoint or method. (POST /api/v1/chat/completions)"}%

curl http://localhost:1234/api/v1/chat \ -H "Content-Type: application/json" \ -d '{ "model": "qwen/qwen2.5-vl-7b", "system_prompt": "You answer only in rhymes.", "input": "What is your favorite color?" }' { "model_instance_id": "qwen/qwen2.5-vl-7b", "output": [ { "type": "message", "content": "My favorite hue, it's a shade so bright,\nA vibrant red, bold and not shy.\nIt's the color of love, passion, and might,\nIn my heart, it always takes flight.\n\nSo when you ask what I adore most,\nThis fiery red, it's my choice." } ], "stats": { "input_tokens": 26, "total_output_tokens": 60, "reasoning_output_tokens": 0, "tokens_per_second": 54.93640195866585, "time_to_first_token_seconds": 0.149 }, "response_id": "resp_3632f5996c12532b444159732a05489620367c967ac516a0" }%