github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #6646] POST /v1/chat/completions returns 404 not 400 for model not found #66219

New Issue
Closed
opened 2026-05-04 00:55:36 -05:00 by GiteaMirror · 3 comments
GiteaMirror commented 2026-05-04 00:55:36 -05:00
Owner
Copy Link

Originally created by @codefromthecrypt on GitHub (Sep 5, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/6646

What is the issue?

POST /v1/chat/completions returns 404 not 400 for model not found. Semantically, the better code here is 400, as it is an invalid argument on a correct route. Using 404 messages on a route that exists is confusing and had me doubting if the routes were mounted or not.

This seems to be the same for other routes. Basically, I suggest any code that looks like this:

			c.JSON(http.StatusNotFound, gin.H{"error": fmt.Sprintf("model '%s' not found", req.Model)})

switch it to bad request

			c.JSON(http.StatusBadRequest, gin.H{"error": fmt.Sprintf("model '%s' not found", req.Model)})

That way, we know the parameter was wrong vs not being able to find the route. I can raise a PR if you like

OS

Linux

GPU

Apple

CPU

No response

Ollama version

0.3.9

Originally created by @codefromthecrypt on GitHub (Sep 5, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/6646 ### What is the issue? POST /v1/chat/completions returns 404 not 400 for model not found. Semantically, the better code here is 400, as it is an invalid argument on a correct route. Using 404 messages on a route that exists is confusing and had me doubting if the routes were mounted or not. This seems to be the same for other routes. Basically, I suggest any code that looks like this: ```go c.JSON(http.StatusNotFound, gin.H{"error": fmt.Sprintf("model '%s' not found", req.Model)}) ``` switch it to bad request ```go c.JSON(http.StatusBadRequest, gin.H{"error": fmt.Sprintf("model '%s' not found", req.Model)}) ``` That way, we know the parameter was wrong vs not being able to find the route. I can raise a PR if you like ### OS Linux ### GPU Apple ### CPU _No response_ ### Ollama version 0.3.9
GiteaMirror added the bug label 2026-05-04 00:55:37 -05:00
GiteaMirror commented 2026-05-04 00:55:51 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Sep 5, 2024):

Thanks @codefromthecrypt. A PR would be awesome!

<!-- gh-comment-id:2330428576 --> @jmorganca commented on GitHub (Sep 5, 2024): Thanks @codefromthecrypt. A PR would be awesome!
GiteaMirror commented 2026-05-04 00:55:56 -05:00
Author
Owner
Copy Link

@codefromthecrypt commented on GitHub (Sep 5, 2024):

actually, I double checked.. openai uses 404 for model not found, surprising to me. So, we have a choice of using 400 only in our endpoints, or using 404 because openai does, for ollama endpoints. preferences?

$ curl -v https://api.openai.com/v1/chat/completions \
>                                 -H "Content-Type: application/json" \
>                                 -H "Authorization: Bearer $OPENAI_API_KEY" \
>                                 -d '{
>                                   "model": "gpt-4l",
>                                   "messages": [
>                                     {
>                                       "role": "user",
>                                       "content": [
>                                         {
>                                           "type": "text",
>                                           "text": "hello?"
>                                         }
>                                       ]
>                                     }
>                                   ],
>                                   "max_tokens": 10
>                                 }'
--snip--
< HTTP/2 404 
< date: Thu, 05 Sep 2024 01:53:03 GMT
< content-type: application/json; charset=utf-8
< content-length: 212
--snip--
{
    "error": {
        "message": "The model `gpt-4l` does not exist or you do not have access to it.",
        "type": "invalid_request_error",
        "param": null,
        "code": "model_not_found"
    }
}
<!-- gh-comment-id:2330450112 --> @codefromthecrypt commented on GitHub (Sep 5, 2024): actually, I double checked.. openai uses 404 for model not found, surprising to me. So, we have a choice of using 400 only in our endpoints, or using 404 because openai does, for ollama endpoints. preferences? ```bash $ curl -v https://api.openai.com/v1/chat/completions \ > -H "Content-Type: application/json" \ > -H "Authorization: Bearer $OPENAI_API_KEY" \ > -d '{ > "model": "gpt-4l", > "messages": [ > { > "role": "user", > "content": [ > { > "type": "text", > "text": "hello?" > } > ] > } > ], > "max_tokens": 10 > }' --snip-- < HTTP/2 404 < date: Thu, 05 Sep 2024 01:53:03 GMT < content-type: application/json; charset=utf-8 < content-length: 212 --snip-- { "error": { "message": "The model `gpt-4l` does not exist or you do not have access to it.", "type": "invalid_request_error", "param": null, "code": "model_not_found" } } ```
GiteaMirror commented 2026-05-04 00:56:02 -05:00
Author
Owner
Copy Link

@pdevine commented on GitHub (Sep 9, 2024):

I would be worried about changing this given other people might be expecting 404 and not 400 in their own applications/scripts. Given there is precedent, let's leave this as it is.

<!-- gh-comment-id:2339255999 --> @pdevine commented on GitHub (Sep 9, 2024): I would be worried about changing this given other people might be expecting 404 and not 400 in their own applications/scripts. Given there is precedent, let's leave this as it is.
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#66219
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?