[GH-ISSUE #22861] issue: OWUI Ollama proxy does not support anthropic compatibility endpoints #58495

New Issue

GiteaMirror · 2026-05-05T23:17:25-05:00

GiteaMirror commented

2026-05-05 23:17:25 -05:00

Originally created by @Umutayb on GitHub (Mar 19, 2026).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/22861

Check Existing Issues

I have searched for any existing and/or related issues.
I have searched for any existing and/or related discussions.
I have also searched in the CLOSED issues AND CLOSED discussions and found no related items (your issue might already be addressed on the development branch!).
I am using the latest version of Open WebUI.

Installation Method

Docker

Open WebUI Version

0.8.10

Ollama Version (if applicable)

0.18.2

Operating System

MacOS, Ubuntu

Browser (if applicable)

No response

Confirmation

I have read and followed all instructions in README.md.
I am using the latest version of both Open WebUI and Ollama.
I have included the browser console logs.
I have included the Docker container logs.
I have provided every relevant configuration, setting, and environment variable used in my setup.
I have clearly listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc).
I have documented step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation. My steps:
Start with the initial platform/version/OS and dependencies used,
Specify exact install/launch/configure commands,
List URLs visited, user input (incl. example values/emails/passwords if needed),
Describe all options and toggles enabled or changed,
Include any files or environmental changes,
Identify the expected and actual result at each stage,
Ensure any reasonably skilled user can follow and hit the same issue.

Expected Behavior

The ollama proxy api spec(request validations) should match the description @ https://docs.ollama.com/api/anthropic-compatibility

curl -X POST "http://localhost:11435/v1/messages" \ -H "Content-Type: application/json" \ -d '{ "model": "qwen3.5:35b", "max_tokens": 1024, "messages": [ { "role": "user", "content": "Hello!" } ] }' {"id":"msg_f639a3a98b1fb905507e9704","type":"message","role":"assistant","model":"qwen3.5:35b","content":[{"type":"thinking","thinking":"Thinking Process:\n\n1. **Analyze the Input:**\n * Input: \"Hello!\"\n * Intent: Greeting.\n * Tone: Friendly, casual.\n * Context: Start of a conversation.\n\n2. **Determine the appropriate response:**\n * Acknowledge the greeting.\n * Offer assistance.\n * Maintain a friendly and helpful tone.\n * Keep it concise but open-ended.\n\n3. **Draft potential responses:**\n * *Option 1:* Hello! How can I help you today?\n * *Option 2:* Hi there! Hope you're having a great day. What's on your mind?\n * *Option 3:* Hello! Nice to meet you. How can I assist you?\n * *Option 4:* Hey! How's it going?\n\n4. **Select the best option:** Option 1 or 2 is usually best for an AI assistant. Let's go with a friendly greeting + offer of help.\n\n5. **Refine the response:** \"Hello! How can I help you today?\" is clear and direct. Let's add a bit more warmth. \"Hello! It's nice to meet you. How can I assist you today?\"\n\n6. **Final Polish:** Keep it simple. \"Hello! How can I help you today?\" is standard and effective. Or \"Hi there! How's your day going?\" Let's combine friendliness with utility.\n\n *Selected:* \"Hello! How can I help you today?\" (Simple, effective).\n *Alternative:* \"Hi! It's great to hear from you. What can I do for you today?\"\n\n Let's go with a warm, friendly greeting that invites further interaction.\n\n7. **Final Output Generation:** \"Hello! How can I help you today?\" (Or similar). I'll make it slightly more engaging.\n\n \"Hello! It's great to hear from you. How can I assist you today?\"\n\n Actually, keeping it punchy is good. \"Hello! How are you doing today? Is there something specific you'd like to chat about or get help with?\"\n\n Let's settle on: \"Hello! How can I help you today?\" (Safe, standard).\n Wait, I can be a bit more conversational.\n \"Hello! 👋 How's your day going? Is there anything I can help you with?\"\n\n Okay, final decision: Friendly greeting + offer of help.\n\n *Draft:* \"Hello! How can I assist you today?\"\n\n *Revised:* \"Hello! It's nice to meet you. How can I help you today?\"\n\n Let's go with a balanced approach.\n\n *Final Choice:* \"Hello! How can I help you today?\" (Simple and direct).\n Actually, adding an emoji or a bit of warmth is nice. \"Hello! 👋 How can I help you today?\"\n\n Let's just output the text.\n\n *Self-Correction during thought:* The user just said \"Hello!\". I should respond in kind.\n\n Response: \"Hello! How can I help you today?\"\n\n Wait, I should check if there are any specific constraints. None.\n\n Okay, ready to output.cw"},{"type":"text","text":"Hello! 👋 How can I help you today?"}],"stop_reason":"end_turn","usage":{"input_tokens":12,"output_tokens":723}}

Actual Behavior

api responds with 405 method not allowed

curl -X POST "http://localhost:3000/ollama/v1/messages" \ -H "Content-Type: application/json" \ -d '{ "model": "qwen3.5:35b", "max_tokens": 1024, "messages": [ { "role": "user", "content": "Hello!" } ] }' {"detail":"Method Not Allowed"}

Steps to Reproduce

Start OWUI via docker.
use /ollama proxy to call ollama api for Anthropic compatibility as described in https://docs.ollama.com/api/anthropic-compatibility.

Logs & Screenshots

2026-03-19 16:57:52.250 | INFO | uvicorn.protocols.http.httptools_impl:send:483 - *********** - "POST /ollama/v1/messages HTTP/1.1" 405

Additional Information

No response

Originally created by @Umutayb on GitHub (Mar 19, 2026). Original GitHub issue: https://github.com/open-webui/open-webui/issues/22861 ### Check Existing Issues - [x] I have searched for any existing and/or related issues. - [x] I have searched for any existing and/or related discussions. - [x] I have also searched in the CLOSED issues AND CLOSED discussions and found no related items (your issue might already be addressed on the development branch!). - [x] I am using the latest version of Open WebUI. ### Installation Method Docker ### Open WebUI Version 0.8.10 ### Ollama Version (if applicable) 0.18.2 ### Operating System MacOS, Ubuntu ### Browser (if applicable) _No response_ ### Confirmation - [x] I have read and followed all instructions in `README.md`. - [x] I am using the latest version of **both** Open WebUI and Ollama. - [x] I have included the browser console logs. - [x] I have included the Docker container logs. - [x] I have **provided every relevant configuration, setting, and environment variable used in my setup.** - [x] I have clearly **listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup** (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc). - [x] I have documented **step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation**. My steps: - Start with the initial platform/version/OS and dependencies used, - Specify exact install/launch/configure commands, - List URLs visited, user input (incl. example values/emails/passwords if needed), - Describe all options and toggles enabled or changed, - Include any files or environmental changes, - Identify the expected and actual result at each stage, - Ensure any reasonably skilled user can follow and hit the same issue. ### Expected Behavior The ollama proxy api spec(request validations) should match the description @ https://docs.ollama.com/api/anthropic-compatibility `curl -X POST "http://localhost:11435/v1/messages" \ -H "Content-Type: application/json" \ -d '{ "model": "qwen3.5:35b", "max_tokens": 1024, "messages": [ { "role": "user", "content": "Hello!" } ] }' {"id":"msg_f639a3a98b1fb905507e9704","type":"message","role":"assistant","model":"qwen3.5:35b","content":[{"type":"thinking","thinking":"Thinking Process:\n\n1. **Analyze the Input:**\n * Input: \"Hello!\"\n * Intent: Greeting.\n * Tone: Friendly, casual.\n * Context: Start of a conversation.\n\n2. **Determine the appropriate response:**\n * Acknowledge the greeting.\n * Offer assistance.\n * Maintain a friendly and helpful tone.\n * Keep it concise but open-ended.\n\n3. **Draft potential responses:**\n * *Option 1:* Hello! How can I help you today?\n * *Option 2:* Hi there! Hope you're having a great day. What's on your mind?\n * *Option 3:* Hello! Nice to meet you. How can I assist you?\n * *Option 4:* Hey! How's it going?\n\n4. **Select the best option:** Option 1 or 2 is usually best for an AI assistant. Let's go with a friendly greeting + offer of help.\n\n5. **Refine the response:** \"Hello! How can I help you today?\" is clear and direct. Let's add a bit more warmth. \"Hello! It's nice to meet you. How can I assist you today?\"\n\n6. **Final Polish:** Keep it simple. \"Hello! How can I help you today?\" is standard and effective. Or \"Hi there! How's your day going?\" Let's combine friendliness with utility.\n\n *Selected:* \"Hello! How can I help you today?\" (Simple, effective).\n *Alternative:* \"Hi! It's great to hear from you. What can I do for you today?\"\n\n Let's go with a warm, friendly greeting that invites further interaction.\n\n7. **Final Output Generation:** \"Hello! How can I help you today?\" (Or similar). I'll make it slightly more engaging.\n\n \"Hello! It's great to hear from you. How can I assist you today?\"\n\n Actually, keeping it punchy is good. \"Hello! How are you doing today? Is there something specific you'd like to chat about or get help with?\"\n\n Let's settle on: \"Hello! How can I help you today?\" (Safe, standard).\n Wait, I can be a bit more conversational.\n \"Hello! 👋 How's your day going? Is there anything I can help you with?\"\n\n Okay, final decision: Friendly greeting + offer of help.\n\n *Draft:* \"Hello! How can I assist you today?\"\n\n *Revised:* \"Hello! It's nice to meet you. How can I help you today?\"\n\n Let's go with a balanced approach.\n\n *Final Choice:* \"Hello! How can I help you today?\" (Simple and direct).\n Actually, adding an emoji or a bit of warmth is nice. \"Hello! 👋 How can I help you today?\"\n\n Let's just output the text.\n\n *Self-Correction during thought:* The user just said \"Hello!\". I should respond in kind.\n\n Response: \"Hello! How can I help you today?\"\n\n Wait, I should check if there are any specific constraints. None.\n\n Okay, ready to output.cw"},{"type":"text","text":"Hello! 👋 How can I help you today?"}],"stop_reason":"end_turn","usage":{"input_tokens":12,"output_tokens":723}}` ### Actual Behavior api responds with 405 method not allowed `curl -X POST "http://localhost:3000/ollama/v1/messages" \ -H "Content-Type: application/json" \ -d '{ "model": "qwen3.5:35b", "max_tokens": 1024, "messages": [ { "role": "user", "content": "Hello!" } ] }' {"detail":"Method Not Allowed"}` ### Steps to Reproduce Start OWUI via docker. use /ollama proxy to call ollama api for `Anthropic compatibility` as described in https://docs.ollama.com/api/anthropic-compatibility. ### Logs & Screenshots `2026-03-19 16:57:52.250 | INFO | uvicorn.protocols.http.httptools_impl:send:483 - *********** - "POST /ollama/v1/messages HTTP/1.1" 405 ` ### Additional Information _No response_

GiteaMirror added the bug label 2026-05-05 23:17:25 -05:00

GiteaMirror closed this issue

2026-05-05 23:17:28 -05:00

GiteaMirror commented

2026-05-05 23:17:30 -05:00

@tjbck commented on GitHub (Mar 20, 2026):

Addressed in dev.

@tjbck commented on GitHub (Mar 20, 2026): Addressed in dev.

GiteaMirror commented

2026-05-05 23:17:32 -05:00

@Umutayb commented on GitHub (Mar 21, 2026):

Would be nice to know the commit id to track it @tjbck