GitHub Copilot Agent Mode Unknown Tokenizer Error for Custom OpenAI-Compatible Models

What this error means

Unknown tokenizer: undefined — Copilot Agent Mode fails for custom OpenAI-compatible models is a GitHub Copilot failure pattern reported for developers trying to fix github copilot agent mode failing with ‘unknown tokenizer: undefined’ for custom openai-compatible models. Based on the imported evidence, treat this as a tool-specific troubleshooting page rather than a generic API error.

Why this happens

GitHub issue #314419 (37 comments) on microsoft/vscode reports Copilot Agent Mode failing with ‘Unknown tokenizer: undefined’ for custom OpenAI-compatible models via customoai provider. Normal chat completions succeed — issue is specific to Agent Mode. Affects VS Code Insiders 1.120.0 + Copilot 0.48.x.

Common causes

• GitHub Copilot Agent Mode fails with ‘Unknown tokenizer: undefined’ when using custom OpenAI-compatible models (configured via chatLanguageModels.json), while normal chat completions work fine. Affects VS Code Insiders 1.120.0 users running custom LLM providers through local proxies.
• GitHub issue #314419 (37 comments) on microsoft/vscode reports Copilot Agent Mode failing with ‘Unknown tokenizer: undefined’ for custom OpenAI-compatible models via customoai provider. Normal chat completions succeed — issue is specific to Agent Mode. Affects VS Code Insiders 1.120.0 + Copilot 0.48.x.

Quick fixes

1. Confirm the exact error signature matches Unknown tokenizer: undefined — Copilot Agent Mode fails for custom OpenAI-compatible models.
2. Check the GitHub Copilot account, local tool state, and provider configuration involved in the failing workflow.
3. Compare the failing environment with a known working setup, then change one configuration value at a time.

Platform/tool-specific checks

• Verify the command, editor, extension, or API client that produced the error.
• Compare local settings with CI, deployment, or editor-level settings when the error appears in only one environment.
• Avoid deleting credentials, local model data, or project settings until the failing scope is clear.

Step-by-step troubleshooting

1. Capture the exact error message and the command, editor action, or request that triggered it.
2. Check whether the failure is account/auth, quota/rate, model/provider, local runtime, or deployment configuration.
3. Review the source evidence below and compare it with your environment.
4. Apply one change at a time and rerun the smallest failing action.
5. Keep the working fix documented for the team or deployment environment.

How to prevent it

• Keep provider/tool configuration documented.
• Record non-secret diagnostics such as tool version, provider name, model name, and command path.
• Add a lightweight check before CI or production workflows depend on the tool.

Sources checked

• https://github.com/microsoft/vscode/issues/314419

Evidence note: GitHub issue #314419 (37 comments) on microsoft/vscode reports Copilot Agent Mode failing with ‘Unknown tokenizer: undefined’ for custom OpenAI-compatible models via customoai provider. Normal chat completions succeed — issue is specific to Agent Mode. Affects VS Code Insiders 1.120.0 + Copilot 0.48.x.

Related errors

• Copilot Agent Mode custom model configuration error
• VS Code custom OpenAI provider tokenizer issue
• GitHub Copilot chatLanguageModels.json Agent Mode bug

FAQ

What should I check first?

Start with the exact Unknown tokenizer: undefined — Copilot Agent Mode fails for custom OpenAI-compatible models text and the smallest action that reproduces it.

Can I ignore this error?

No. Treat it as a failed GitHub Copilot workflow until the root cause is understood.

Is this guaranteed to have one fix?

No. The imported evidence supports the troubleshooting path above, but tool behavior can vary by account, plan, version, provider, and local configuration.

How do I know the fix worked?

Rerun the same command, editor action, or request. The fix is working when that action completes without Unknown tokenizer: undefined — Copilot Agent Mode fails for custom OpenAI-compatible models.