[BUG] VS Code extension ignores third-party provider environment variables during authentication check

Found 3 possible duplicate issues:

1. https://github.com/anthropics/claude-code/issues/11937
2. https://github.com/anthropics/claude-code/issues/9136
3. https://github.com/anthropics/claude-code/issues/11027

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

Might be related, as I know the #11937 post I shared, indicated I got it working with CLI locally, but not in Azure Pipelines/CI, so following in case there's some related issue that in different tooling or environment is causing it to fail.

hello, any updates on this?

Agree, any updates on this? I have spent a whole lot of time trying to get this work, but I am starting to think this isn't actually a feature anymore.

At least the selected model id should be clearly displayed somewhere like what it does in CLI, currently it's very confused which one is selected. I have to ask the model who you are.

This issue has been inactive for 30 days. If the issue is still occurring, please comment to let us know. Otherwise, this issue will be automatically closed in 30 days for housekeeping purposes.

I left claude extension.

I'm now using GPT codex extension instead.

The bug is probably still happening.

The idea was to use anthropic/vertex API in the claude extension, however, the setup gas never worked.

This issue has been automatically closed due to 60 days of inactivity. If you're still experiencing this issue, please open a new issue with updated information.

This issue was closed incorrectly despite recent human comments. This behavior of the bot is reported at https://github.com/anthropics/claude-code/issues/16497. Please upvote that issue, so maybe it gets noticed.

Even though this issue was closed as “not planned”, is there any chance this behavior could be revisited in the future roadmap?

This problem still seems to affect users in similar scenarios, and it would be helpful to know whether a fix or redesign is explicitly out of scope or just not prioritized at the moment.

Any clarification on long-term plans would be greatly appreciated.

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:

``bash
ollama serve
``

1. Model pulled:

``bash
ollama pull kimi-k2.5:latest
``

1. Verify Ollama API is accessible:

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

This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.