The specified Ollama embedding model was not found. Please ensure the model is installed locally or select a different embedding model in the configuration.

[zhenaiwork]

2025-07-02 01:29:31,310 - INFO - api.websocket_wiki - websocket_wiki.py:69 - Request size: 7782 tokens
2025-07-02 01:29:31,311 - WARNING - api.ollama_patch - ollama_patch.py:56 - Could not connect to Ollama to check models: No connection adapters were found for '0.0.0.0:11434/api/tags'
2025-07-02 01:29:31,311 - ERROR - api.websocket_wiki - websocket_wiki.py:111 - Error preparing retriever: Ollama model 'nomic-embed-text' not found. Please run 'ollama pull nomic-embed-text' to install it.

在点击生成后会出现以上报错

这个是我的.env 文件

# No need for API keys when using Ollama locally
PORT=8001
# Optionally, provide OLLAMA_HOST if Ollama is not local
OLLAMA_HOST=http://localhost:11434 # (default: http://localhost:11434)

还有我的embedder.json & ollama

{
  "embedder": {
    "client_class": "OllamaClient",
    "model_kwargs": {
      "model": "nomic-embed-text"
    }
  },
  "retriever": {
    "top_k": 20
  },
  "text_splitter": {
    "split_by": "word",
    "chunk_size": 350,
    "chunk_overlap": 100
  }
}

$ ollama list                         
NAME                                 ID              SIZE      MODIFIED       
nomic-embed-text:latest              0a109f422b47    274 MB    29 minutes ago    
qwen3:1.7b                           8f68893c685c    1.4 GB    3 hours ago       
qwen3:8b                             500a1f067a9f    5.2 GB    6 hours ago       
dengcao/Qwen3-Embedding-8B:Q4_K_M    d3ca2355027f    4.7 GB    14 hours ago      
qwen3:32b                            030ee887880f    20 GB     15 hours ago      
qwen3:latest                         e4b5fd7f8af0    5.2 GB    6 weeks ago  

[soteaw]

It looks like you have OLLAMA_HOST=0.0.0.0 set in your environment variables. Try unsetting it locally before starting the Backend.

[ahmadmujtaba36]

Hello @g60cBQ, I have the same issue. I have the same configuration as @zhenaiwork. Unfortunately, I could not understand your solution to the issue. Can you please elaborate?

[soteaw]

@ahmadmujtaba36 Sure. Your system has an OLLAMA_HOST=0.0.0.0 environment variable. The Ollama server uses this to listen for incoming connections (which is fine). However, deepwiki-open also sees this variable and tries to connect to the address 0.0.0.0, which doesn't work.

The simplest solution is to temporarily remove this variable in your current terminal session. This allows deepwiki-open to use the correct address you've set in .env file.

Here's how to do it before you start the backend:

For Linux

unset OLLAMA_HOST
# Then, start the backend

For Windows PowerShell

Remove-Item Env:OLLAMA_HOST
# Then, start the backend

[kroryan]

@ahmadmujtaba36 Sure. Your system has an OLLAMA_HOST=0.0.0.0 environment variable. The Ollama server uses this to listen for incoming connections (which is fine). However, deepwiki-open also sees this variable and tries to connect to the address 0.0.0.0, which doesn't work.

The simplest solution is to temporarily remove this variable in your current terminal session. This allows deepwiki-open to use the correct address you've set in .env file.

Here's how to do it before you start the backend:

For Linux

unset OLLAMA_HOST
# Then, start the backend

For Windows PowerShell

Remove-Item Env:OLLAMA_HOST
# Then, start the backend

im facing the same issue i did it but still not working ... any idea what i can check now?