Kimi K3 overtakes Fable 5 on Code Arena in just six weeks
arena · x · 2026-07-24
Kimi K3 has moved ahead of Fable 5 on Code Arena just six weeks after Fable’s release, reaching No. 1 in 6 of 7 frontend domains.
The interview clip highlights:
- A 1,679-point score and a 17-place jump from K2.6.
- A data point suggesting Chinese open-weight models may be distilling rapidly.
- A debate on whether this is a permanent shift or just a snapshot.
- A pricing angle: “$3 in, $15 out” is making open source look less like a 10% cost option.
- A strategic question for Anthropic and OpenAI: what do they do next?
Related event: Kimi K3 Tops Code and Design Arenas Amid Distillation Debate(7 posts)→
More from Models
- Kimi K2.8 Preview rolls out: near-K3 coding performance, 1M context for all tiers — teortaxesTex · 2026-09-11
- Looking for a classifier of software engineering task shapes to pick models per task — StewartalsopIII · 2026-09-11
- DeepSeek V4 Pro API to continue after Sept 2026, billing unchanged — teortaxesTex · 2026-09-11
- DeepSeek V4.1 Flash Hits 98% of GPT-6 Astra's Score at 1.4% of the Cost in Third-Party Benchmark — ayushtweetshere · 2026-09-11
- TheZvi Polls: Has Your Coding Model Choice Changed Since Fable 5.1 and Astra? — TheZvi · 2026-09-11
- antirez Weighs In on Anthropic Banning Minors From Using Claude — antirez · 2026-09-11