Context window below 50% but LLM provider error: Error code: 400 - {'error': {'message': 'Invalid request: Your request exceeded model token limit: 262144 (re quested: 269030)'

What version of Kimi Code CLI is running?

kimi, version 1.37.0

Which open platform/subscription were you using?

moonshot.ai

Which model were you using?

kimi-2.5

What platform is your computer?

Linux 6.8.0-110-generic x86_64 x86_64

What issue are you seeing?

LLM provider error: Error code: 400 - {'error': {'message': 'Invalid request: Your request exceeded model token limit: 262144 (re quested: 269030)', 'type': 'invalid_request_error'}} If this persists, run kimi export and send the exported data to support for assistance. Please do not share the exported file pub licly. ✨ again LLM provider error: Error code: 400 - {'error': {'message': 'Invalid request: Your request exceeded model token limit: 262144 (re quested: 313014)', 'type': 'invalid_request_error'}} If this persists, run kimi export and send the exported data to support for assistance. Please do not share the exported file pub licly. LLM provider error: Error code: 400 - {'error': {'message': 'Invalid request: Your request exceeded model token limit: 262144 (re quested: 269053)', 'type': 'invalid_request_error'}} If this persists, run kimi export and send the exported data to support for assistance. Please do not share the exported file pub licly. Exported 7 messages to ~/kimi-export-599a0bb3-20260423-003025.md Note: The exported file may contain sensitive information. Please be cautious when sharing it externally. ✨ now? LLM provider error: Error code: 400 - {'error': {'message': 'Invalid request: Your request exceeded model token limit: 262144 (re quested: 313016)', 'type': 'invalid_request_error'}} If this persists, run kimi export and send the exported data to support for assistance. Please do not share the exported file pub licly.

What steps can reproduce the bug?

dont know

What is the expected behavior?

no matter what I type the error comes again:
LLM provider error: Error code: 400 - {'error': {'message': 'Invalid request: Your request exceeded model token limit: 262144 (re quested: 269030)', 'type': 'invalid_request_error'}} If this persists, run kimi export and send the exported data to support for assistance. Please do not share the exported file pub licly.

Additional information

No response

Start by reproducing the error with Kimi Code CLI 1.37.0 on Linux using moonshot.ai and the kimi-2.5 model. Compare the reported context-window percentage with the provider's requested token count, using the exported conversation only with appropriate care. Done means the excessive-token error is reproduced and its handling or trigger is addressed without recurring on normal input.