Kimi K2.7 Post-Training on 2,400 Coding Tasks Boosts Benchmarks

Kimi's team reported that post-training K2.7 on 2,400 coding tasks lifted DeepSWE from 31% to 48.7% (+17.7pp), with SWE-Bench Pro up about 8 points, showing large gains across coding benchmarks.

2026-10-22 ~ 2026-10-22 · 2 related posts

• Kimi K2.7 post-trained on 2,400 coding tasks lifts DeepSWE from 31% to 48.7% — echen · 2026-10-22
• Kimi K2.7 post-trained on 2,400 coding tasks lifts SWE-Bench Pro to 68.1% — echen · 2026-10-22