[BUG] Extremely long time to first token
Resolved 💬 3 comments Opened Jul 11, 2025 by jannalulu Closed Nov 30, 2025
Environment
- Platform (select one):
- [x] Anthropic API
- [ ] AWS Bedrock
- [ ] Google Vertex AI
- [ ] Other: <!-- specify -->
- Claude CLI version: 1.0.48
- Operating System: Ubuntu 22.04 (WSL)
- Terminal: VSCode
Bug Description
Time to first token is EXTREMELY long, typically between 5-7 minutes. Claude Code would have 0 tokens up/down and the timer would keep increasing.
Steps to Reproduce
- open WSL and start Claude Code with
claude - Ask a simple query, such as
please explain how [insert file here] works.
Expected Behavior
Wait a few seconds, and the token counter changes. Ideally Claude also starts outputting text.
Actual Behavior
Waited 8 minutes before Claude Code token counter changed and it started outputting tokens for a response to a simple question.
Additional Context
<img width="416" height="118" alt="Image" src="https://github.com/user-attachments/assets/f8c762af-f2ff-4b88-a031-22a5976149bf" />
<img width="386" height="113" alt="Image" src="https://github.com/user-attachments/assets/9d745a56-0d04-4e65-bc13-966eccee06df" />
This issue has 3 comments on GitHub. Read the full discussion on GitHub ↗