[BUG] [FIX] VS Code Extension does not recognize ANTHROPIC_BASE_URL and shows login screen

[shaunyogeshwaran]

Preflight Checklist

• I have searched existing issues and this hasn't been reported yet
• This is a single bug report (please file separate reports for different bugs)
• I am using the latest version of Claude Code

What's Wrong?

Fixed issue #8522 by configuring "claudeCode.disableLoginPrompt": true and adding LiteLLM proxy environment variables (LITELLM_PROXY_API_KEY and LITELLM_PROXY_URL) in VS Code settings. This solution bypasses the login screen and routes requests through LiteLLM to custom backends, providing a working alternative to the unsupported ANTHROPIC_BASE_URL feature.

I'd assume this works with other 3rd party providers too

What Should Happen?

Claude Code should respect environment variables for custom backends and bypass the Anthropic login screen when configured with alternative LLM providers.

Error Messages/Logs

Steps to Reproduce

1. Install Claude Code 2.0.0+
2. Set ANTHROPIC_BASE_URL in environment or configuration file
3. Open VS Code and try to use Claude Code
4. Observe that Claude Code ignores the custom URL and displays Anthropic login screen
5. Unable to proceed with custom LLM backend configuration

Claude Model

Not sure / Multiple models

Is this a regression?

No, this never worked

Last Working Version

No response

Claude Code Version

2.1.17

Platform

Other

Operating System

macOS

Terminal/Shell

iTerm2

Additional Information

No response

[github-actions]

Found 3 possible duplicate issues:

1. [BUG] New VS Code Extension does not recognize ANTHROPIC_BASE_URL and shows login screen #8522
2. [BUG] [VS Code Extension] LLM Gateway/Custom ANTHROPIC_BASE_URL not working in v2.0.0+ #9010
3. [BUG] VS Code extension ignores third-party provider environment variables during authentication check #12361

This issue will be automatically closed as a duplicate in 3 days.

• If your issue is a duplicate, please close it and 👍 the existing issue instead
• To prevent auto-closure, add a comment or 👎 this comment

🤖 Generated with Claude Code

[homerokzam]

Configuration Guide: Using Ollama Local Models with Claude Code VSCode Extension

Summary

This issue documents the working configuration for using Ollama-hosted local models with the Claude Code VSCode Extension, addressing the setup challenges reported in #20271 and #12361.

Working Configuration

To use Ollama local models, split the configuration between two files:

1. .vscode/settings.json - Model Selection

{
  "claudeCode.selectedModel": "kimi-k2.5:cloud"
}

2. .claude/settings.local.json - Ollama Environment

{
  "env": {
    "ANTHROPIC_AUTH_TOKEN": "ollama",
    "ANTHROPIC_API_KEY": "",
    "ANTHROPIC_MODEL": "kimi-k2.5:latest",
    "ANTHROPIC_BASE_URL": "http://localhost:11434"
  }
}

Prerequisites

1. Ollama installed and running:

   ollama serve

2. Model pulled:

   ollama pull kimi-k2.5:latest

3. Verify Ollama API is accessible:

   curl http://localhost:11434/api/tags

[github-actions]

Closing for now — inactive for too long. Please open a new issue if this is still relevant.