MoonshotAI / MoonshotAI/kimi-code

Error with models not supporting prompt cache

Open
#2,166 6 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug
Dominant language
TypeScript
Stars
7.5k
Forks
1.2k
Avg merge
11h 53m
Merged PRs (30d)
350

Description

What version of Kimi Code is running?

0.29.1

Which open platform/subscription were you using?

nvidia nim

Which model were you using?

nemotron, minimax

What platform is your computer?

windows, wsl

What issue are you seeing?

This was working fine until the latest update today. I believe Nvidia NIM doesn't support this but I have been using their models for weeks with kimi code. This week's update broke it somehow. Is there a setting I can use to disable it? Thanks.

Model API error
400 Validation: Unsupported parameter(s): prompt_cache_key

What steps can reproduce the bug?

use 3rd party models without prompt_cache support will error

What is the expected behavior?

No response

Additional information

No response

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reproducing the 400 error with the listed NVIDIA NIM models and trace where the CLI adds prompt_cache_key to model API requests. The fix is done when models without prompt-cache support no longer receive the unsupported parameter while compatible models continue to work, with regression coverage for both cases.

Written by the indexing model from the issue text.

Assessment

Tech stack
typescript
Domain
api, cli
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Active
Clarity
Mostly clear
Newbie friendliness
66/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.