in vscode, CONTINUE Can not connect to a public URL when using remote ollama

[PeterHenTai]

Before submitting your bug report

• I believe this is a bug. I'll try to join the Continue Discord for questions
• I'm not able to find an open issue that reports the same bug
• I've seen the troubleshooting guide on the Continue Docs

Relevant environment info

- OS:Linux
- Continue version:0.8.68
- IDE version:1.97.0
- Model:deepseek-coder:1.3b
- config:
  
{
  "models": [
    {
      "apiBasse": "https://41245a9c.r8.cpolar.top",
      "model": "deepseek-coder:1.3b",
      "title": "deepseek_yuyx",
      "provider": "ollama"
    }
  ],
  "tabAutocompleteModel": {
    "apiBasse": "https://41245a9c.r8.cpolar.top",
    "model": "deepseek-coder:1.3b",
    "title": "deepseek_yuyx",
    "provider": "ollama"
  },
  "contextProviders": [
    {
      "name": "code",
      "params": {}
    },
    {
      "name": "docs",
      "params": {}
    },
    {
      "name": "diff",
      "params": {}
    },
    {
      "name": "terminal",
      "params": {}
    },
    {
      "name": "problems",
      "params": {}
    },
    {
      "name": "folder",
      "params": {}
    },
    {
      "name": "codebase",
      "params": {}
    }
  ],
  "slashCommands": [
    {
      "name": "share",
      "description": "Export the current chat session to markdown"
    },
    {
      "name": "cmd",
      "description": "Generate a shell command"
    },
    {
      "name": "commit",
      "description": "Generate a git commit message"
    }
  ]
}

Description

My colleague ran the DeepSeek model on his computer using Olama and deployed the model to the public network using a private network penetration tool. After configuring apiBase using the continue plugin in VSCode, I was unable to use it and encountered the error 'Unable to connect to local Olama instance' Olama may not be running. 'However, if I use the same config. json on Windows, it can be used normally. It looks like a bug in the Linux version

To reproduce

1. install continue plugin in VScode
2. edit config.json as this bug
3. click on enter
4. see Unable to connect to local Olama instance' Olama may not be running

Log output

[x2x4com]

Maybe you can try add apiKey on config and reconfig your server(add some kind of proxy or api gateway for you ollama, simple check nginx with auth mod) to auth with Authorization header(Not sure will work, need test)

[gerrith3]

Make sure you can connect to his version of ollama e.g. with curl his.ip.address:his-port to make sure ollama is running and accessible. There isn't enough in the defect report to be sure that ollama is actually serving your IP. If that isn't working, make sure to get that debugged first... Also, isn't apiBasse supposed to be apiBase ?

[philippeflorent]

where does one putt these settings ? my config file has so not the same format

name: Local Assistant
version: 1.0.0
schema: v1
models:
  - name: codellama:7b-instruct
    provider: ollama
    model: codellama:7b-instruct
    apiBasse: 172.29.234.47:11434 >>>> "apiBasse" not accepted by vscode
    roles:
      - chat
      - edit
      - apply
      - embed

also , what is an api basse ? did you mean base ?

[gerrith3]

So the top of this issue is the old format - config.json; your format is the newer, config.yaml. See: https://docs.continue.dev/yaml-migration for more details.