github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-16 21:25:46 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #5830] OpenAI endpoint gives 404 #29392

New Issue
Closed
opened 2026-04-22 08:14:35 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-04-22 08:14:35 -05:00
Owner
Copy Link

Originally created by @defaultsecurity on GitHub (Jul 21, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/5830

What is the issue?

Otherwise Ollama is working. I'm not sure what to do.

OS

Linux

GPU

Nvidia

CPU

AMD

Ollama version

0.2.7

Originally created by @defaultsecurity on GitHub (Jul 21, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/5830 ### What is the issue? - http://localhost:11434/v1/chat/completions (gives 404) - http://localhost:11434 (shows ollama is running) Otherwise Ollama is working. I'm not sure what to do. ### OS Linux ### GPU Nvidia ### CPU AMD ### Ollama version 0.2.7
GiteaMirror added the bug label 2026-04-22 08:14:35 -05:00
GiteaMirror commented 2026-04-22 08:14:37 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jul 21, 2024):

If you're using curl or wget to test the endpoint, be aware that by default these tools send a GET and /v1/chat/completions requires a POST.

$ curl -D - http://localhost:11434/v1/chat/completions
HTTP/1.1 404 Not Found
Content-Type: text/plain
Date: Sun, 21 Jul 2024 14:31:09 GMT
Content-Length: 18

404 page not found

$ curl -D - -X POST http://localhost:11434/v1/chat/completions
HTTP/1.1 400 Bad Request
Content-Type: application/json; charset=utf-8
Date: Sun, 21 Jul 2024 14:31:44 GMT
Content-Length: 83

{"error":{"message":"EOF","type":"invalid_request_error","param":null,"code":null}}

curl will automatically use POST when you supply data (-d, --data):

$ curl -D - http://localhost:11434/v1/chat/completions -d '{"model": "gemma2", "messages": [{"role": "user", "content": "hello"}], "stream": false}'
HTTP/1.1 200 OK
Content-Type: application/json
Date: Sun, 21 Jul 2024 14:36:54 GMT
Content-Length: 326

{"id":"chatcmpl-385","object":"chat.completion","created":1721572614,"model":"gemma2","system_fingerprint":"fp_ollama","choices":[{"index":0,"message":{"role":"assistant","content":"Hello! 👋\n\nHow can I help you today? 😊"},"finish_reason":"stop"}],"usage":{"prompt_tokens":10,"completion_tokens":13,"total_tokens":23}}
<!-- gh-comment-id:2241632971 --> @rick-github commented on GitHub (Jul 21, 2024): If you're using `curl` or `wget` to test the endpoint, be aware that by default these tools send a `GET` and /v1/chat/completions requires a `POST`. ``` $ curl -D - http://localhost:11434/v1/chat/completions HTTP/1.1 404 Not Found Content-Type: text/plain Date: Sun, 21 Jul 2024 14:31:09 GMT Content-Length: 18 404 page not found ``` ``` $ curl -D - -X POST http://localhost:11434/v1/chat/completions HTTP/1.1 400 Bad Request Content-Type: application/json; charset=utf-8 Date: Sun, 21 Jul 2024 14:31:44 GMT Content-Length: 83 {"error":{"message":"EOF","type":"invalid_request_error","param":null,"code":null}} ``` `curl` will automatically use `POST` when you supply data (-d, --data): ``` $ curl -D - http://localhost:11434/v1/chat/completions -d '{"model": "gemma2", "messages": [{"role": "user", "content": "hello"}], "stream": false}' HTTP/1.1 200 OK Content-Type: application/json Date: Sun, 21 Jul 2024 14:36:54 GMT Content-Length: 326 {"id":"chatcmpl-385","object":"chat.completion","created":1721572614,"model":"gemma2","system_fingerprint":"fp_ollama","choices":[{"index":0,"message":{"role":"assistant","content":"Hello! 👋\n\nHow can I help you today? 😊"},"finish_reason":"stop"}],"usage":{"prompt_tokens":10,"completion_tokens":13,"total_tokens":23}} ```
GiteaMirror commented 2026-04-22 08:14:38 -05:00
Author
Owner
Copy Link

@defaultsecurity commented on GitHub (Jul 22, 2024):

You're aboslutely right. Thank you for the information.

<!-- gh-comment-id:2242172241 --> @defaultsecurity commented on GitHub (Jul 22, 2024): You're aboslutely right. Thank you for the information.
GiteaMirror commented 2026-04-22 08:14:39 -05:00
Author
Owner
Copy Link

@gamersalpha commented on GitHub (Jan 30, 2026):

I'm experiencing the exact same issue on a fresh Ollama installation.

Environment:

  • OS: Debian Trixie (testing)
  • Ollama version: Latest (installed today via curl -fsSL https://ollama.com/install.sh | sh)
  • CPU-only setup (no GPU)
  • Ollama running as systemd service

What works:

curl http://127.0.0.1:11434/api/generate -d '{
  "model": "mistral:7b",
  "prompt": "Hello",
  "stream": false
}'
# ✅ Returns valid response in ~4.8 seconds

What doesn't work:

curl http://127.0.0.1:11434/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mistral:7b",
    "messages": [{"role": "user", "content": "Hello"}],
    "stream": false
  }'
# ❌ Hangs indefinitely, no response, no error

The native Ollama API (/api/generate) works perfectly, but the OpenAI-compatible endpoint (/v1/chat/completions) doesn't respond at all - it just hangs without returning any HTTP status code.

I tried using litellm as a proxy with --drop_params flag but that only revealed that the /v1 endpoint itself isn't responding to requests.

This makes Ollama incompatible with tools that require OpenAI-compatible APIs (like OpenClaw/Clawdbot).

Is the /v1 endpoint supposed to work out of the box, or does it require specific configuration?

<!-- gh-comment-id:3825613434 --> @gamersalpha commented on GitHub (Jan 30, 2026): I'm experiencing the exact same issue on a fresh Ollama installation. **Environment:** - OS: Debian Trixie (testing) - Ollama version: Latest (installed today via `curl -fsSL https://ollama.com/install.sh | sh`) - CPU-only setup (no GPU) - Ollama running as systemd service **What works:** ```bash curl http://127.0.0.1:11434/api/generate -d '{ "model": "mistral:7b", "prompt": "Hello", "stream": false }' # ✅ Returns valid response in ~4.8 seconds ``` **What doesn't work:** ```bash curl http://127.0.0.1:11434/v1/chat/completions \ -H "Content-Type: application/json" \ -d '{ "model": "mistral:7b", "messages": [{"role": "user", "content": "Hello"}], "stream": false }' # ❌ Hangs indefinitely, no response, no error ``` The native Ollama API (`/api/generate`) works perfectly, but the OpenAI-compatible endpoint (`/v1/chat/completions`) doesn't respond at all - it just hangs without returning any HTTP status code. I tried using litellm as a proxy with `--drop_params` flag but that only revealed that the `/v1` endpoint itself isn't responding to requests. This makes Ollama incompatible with tools that require OpenAI-compatible APIs (like OpenClaw/Clawdbot). Is the `/v1` endpoint supposed to work out of the box, or does it require specific configuration?
GiteaMirror commented 2026-04-22 08:14:40 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jan 31, 2026):

It is not the exact same issue. Open a new issue.

<!-- gh-comment-id:3828551311 --> @rick-github commented on GitHub (Jan 31, 2026): It is not the exact same issue. Open a new issue.
GiteaMirror commented 2026-04-22 08:14:42 -05:00
Author
Owner
Copy Link

@carloslockward commented on GitHub (Feb 8, 2026):

@gamersalpha Did you find a solution?

<!-- gh-comment-id:3867969098 --> @carloslockward commented on GitHub (Feb 8, 2026): @gamersalpha Did you find a solution?
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#29392
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?