Bug Report

Installation Method

Installed via VSCode extension continue.dev (v0.8.55)

Environment

• Open WebUI Version: 0.3.35
• Operating System: macOS (arm64)
• Browser: VSCode Extension

Confirmation:

• [x] I have read and followed all the instructions provided in the README.md.
• [x] I am on the latest version of both Open WebUI and continue.dev.
• [x] I have included the browser console logs.
• [x] I have provided the exact steps to reproduce the bug in the "Steps to Reproduce" section below.

Expected Behavior:

The continue.dev integration should work as described in the official documentation.

Actual Behavior:

Multiple errors when trying to use the integration:

• 404 Not Found errors when using /v1 endpoint
• 400 Bad Request errors when using alternative configurations
• Model not found errors with various model name formats

Description

Bug Summary: The continue.dev integration as documented in OpenWebUI docs fails to work with any of the suggested configurations.

Reproduction Details

Steps to Reproduce:

1. Install continue.dev VSCode extension
2. Configure continue.dev according to the documentation:
3. Try to use the model in VSCode

Logs and Screenshots

VSCode Extension Logs:

Error generating autocompletion: Error: HTTP 404 Not Found from https://[redacted]/ollama/v1/completions
{"detail":"Not Found"}

Alternative configuration attempts also failed:

HTTP 400 Bad Request from https://[redacted]/ollama/api/generate
{"detail":"Model 'model-name' was not found"}

Additional Information

• Direct API calls via curl/httpie work correctly

➜  .continue http POST 'https://xxx/ollama/api/generate' \
  'Authorization: Bearer sk-xxx' \
  'Content-Type: application/json' \
  model='granite-code:latest' \
  prompt='Write a simple Hello World in Python'
HTTP/1.1 200 OK
Connection: keep-alive
Content-Type: application/x-ndjson
Date: Tue, 29 Oct 2024 21:26:02 GMT
Set-Cookie: AWSALB=8P0zyD9kIYzUHXV8ig6vINbie62pivqU5YXf0NFz2YSRq42KlUHb4FsLgolVdrUrQc+VlGRldbobs2/hBWm6r+19IYGEtS+Wqb2YaNkZaHo3YoePu65D7rPIUXDX; Expires=Tue, 05 Nov 2024 21:25:59 GMT; Path=/
Set-Cookie: AWSALBCORS=8P0zyD9kIYzUHXV8ig6vINbie62pivqU5YXf0NFz2YSRq42KlUHb4FsLgolVdrUrQc+VlGRldbobs2/hBWm6r+19IYGEtS+Wqb2YaNkZaHo3YoePu65D7rPIUXDX; Expires=Tue, 05 Nov 2024 21:25:59 GMT; Path=/; SameSite=None; Secure
Transfer-Encoding: chunked
server: uvicorn
x-process-time: 3

{"model":"granite-code:latest","created_at":"2024-10-29T21:26:02.293386453Z","response":"print","done":false}
{"model":"granite-code:latest","created_at":"2024-10-29T21:26:02.293400088Z","response":"(\"","done":false}
{"model":"granite-code:latest","created_at":"2024-10-29T21:26:02.298295184Z","response":"Hello","done":false}
{"model":"granite-code:latest","created_at":"2024-10-29T21:26:02.311280454Z","response":",","done":false}
{"model":"granite-code:latest","created_at":"2024-10-29T21:26:02.324085076Z","response":" World","done":false}
{"model":"granite-code:latest","created_at":"2024-10-29T21:26:02.336842177Z","response":"!\")","done":false}
{"model":"granite-code:latest","created_at":"2024-10-29T21:26:02.34973698Z","response":"\n","done":false}
{"model":"granite-code:latest","created_at":"2024-10-29T21:26:02.362694032Z","response":"","done":true,"done_reason":"stop","context":[9927,44,203,2538,312,4281,12000,10896,328,4865,203,203,10963,44,203,1216,440,8279,30,10896,15981,203],"total_duration":2520269464,"load_duration":2403134579,"prompt_eval_count":15,"prompt_eval_duration":25726000,"eval_count":8,"eval_duration":90332000}

• Models are available and listed correctly when querying the API directly

{
    "data": [
        {
            "created": 1730237091,
            "id": "granite-code:latest",
            "object": "model",
            "owned_by": "openai"
        },
        {
            "created": 1730237091,
            "id": "deepseek-coder:latest",
            "object": "model",
            "owned_by": "openai"
        },
        {
            "created": 1730237091,
            "id": "deepseek-coder-v2:16b",
            "object": "model",
            "owned_by": "openai"
        },
        {
            "created": 1730237091,
            "id": "codestral:22b",
            "object": "model",
            "owned_by": "openai"
        },
        {
            "created": 1730237091,
            "id": "starcoder2:3b",
            "object": "model",
            "owned_by": "openai"
        },
        {
            "created": 1730237091,
            "id": "command-r:35b",
            "object": "model",
            "owned_by": "openai"
        },
        {
            "created": 1730237091,
            "id": "gemma2:latest",
            "object": "model",
            "owned_by": "openai"
        }
    ],
    "object": "list"
}

• Documentation might need updating to reflect the correct configuration requirements
• Tried multiple configuration variants including:
  • Different providers (openai/ollama)
  • Different API base paths
  • Different model name formats
  • With/without legacy completions endpoint flag

Note

All steps have been verified and the issue persists across different configuration attempts. The API works when tested directly, suggesting this is specifically a continue.dev integration issue.