429 error "We're receiving too many requests at the moment. Please wait a moment and try again."

Kimi K2
1

Getting this error for over an hour now for api keys and the kimi code program itself. I had recently updated api keys in openclaw and now it doesn’t work at all. I have the Allegretto membership and am well below my daily or even weekly limits.

2

Hi there! Thanks so much for posting this, and for your support as an Allegretto member.

I completely understand the frustration of seeing an error when you know you have a valid membership and remaining quota.

Here is the situation regarding those limits:

  • Concurrency vs. Total Quota: You are likely hitting a concurrency limit (simultaneous requests) rather than your total volume allowance. This is why your dashboard still shows a healthy remaining quota.
  • Rapid Growth: We’ve seen an incredible surge in user enthusiasm recently, which has led to a sharp increase in system load.
  • Scaling & Documentation: To support this growth, we are actively raising these concurrency limits. At the same time, we are updating our documentation to clearly explain these specific limits so the distinction is transparent moving forward.

This is similar to the issue discussed in Error code: 429: We're receiving too many requests at the moment.

Thanks again for bearing with us as we scale!

1 Like
3

Hi there! :waving_hand:

Following up on my message from yesterday — wanted to share that we’ve pushed some updates and clarified our documentation.

Per the Key Advantages section now live at https://www.kimi.com/code/docs/en/:

Key Advantages

  • Seamless Integration: Full compatibility with Kimi Code CLI, Claude Code, and Roo Code, fitting perfectly into your existing CI/CD or local workflows.
  • Elite Performance: Experience blistering output speeds of up to 100 Tokens/s with high stability.
  • Throughput Capacity: A 5-hour token quota supports approximately 300–1,200 API calls, with a maximum concurrency of 30, ensuring uninterrupted operation for complex workloads.

What changed behind the scenes:

  1. Quota Increase: We’ve raised the concurrency allowance for Allegretto members to that maximum concurrency of 30 quoted above — this directly addresses the 429 bottleneck you hit when updating your OpenClaw keys.

  2. Documentation Clarified: The docs now explicitly spell out the 30-request concurrency ceiling and throughput numbers (300–1,200 calls per 5-hour window) so there’s no ambiguity on where limits lie.

  3. Infrastructure: We’ve made incremental capacity improvements and continue working to ensure stability as usage grows.

That concurrency limit of 30 should resolve the “completely stopped working” scenario you experienced. You now have significantly more headroom for simultaneous requests through OpenClaw.

Would you mind giving it another try? If you still hit walls with your specific batch setup, let us know what concurrency level you’re typically running — helps us ensure the new limits align with real-world usage.

Thanks for your patience!