# vLLM reports K3.1 Pro scores 61.2% on Kimi Code Bench 3.0 Long-horizon

> Original title: Long tasks are the point.

- Source: vLLM
- Published: 2026-11-04T16:07:57.000Z
- Bad Influence: https://badinfluence.news/items/l8hpooinhevsbt67vbrmj0eu2
- Original: https://x.com/vllm_project/status/2118012381276291137

## Summary

vLLM reports that Moonshot AI's K3.1 Pro scores 61.2% on the internal Kimi Code Bench 3.0 Long-horizon benchmark, up from K3's 55.95%. The benchmark covers multi-step execution, tool use, and end-to-end engineering. Both coding results were measured with the Claude Code harness at maximum reasoning effort.

## Post

Long tasks are the point.

K3.1 Pro scores 61.2% on Kimi Code Bench 3.0 Long-horizon, up from K3’s 55.95%. This internal benchmark covers multi-step execution, tool use and end-to-end engineering.

Both coding results use the Claude Code harness at maximum reasoning effort.

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHWSxCaObkAAoHcd.jpg&mode=thumb&exp=1794009600&sig=bc2bf313452b3c37)