New Kimi Model K3.1 Spotted on Code Arena
PMinervini · x · 2026-08-22
A new model from Moonshot, likely K3.1, is being tested on the LMSYS Chatbot Arena under the name "korrine" in the coding category. Previously, the K3 model was tested as "kivine" before its launch. Tim Dettmers noted that Zhipu is one of the few labs with slow partial prefill but fast output tok/s; he speculates that if the new Kimi model is 30% faster than GLM 5.3, it might have 300-500B or fewer parameters (active params).
Related event: Mystery model 'korrine' on LM Arena suspected to be Kimi K3.1(4 posts)→
More from Models
- Meta's open-weight Muse Spark Open hits Ollama with 1M-token context and Claude Code support — ollama · 2026-10-16
- 'Opus 5.5' collage demos spark claim: models have taste, no moat — yuwen_lu_ · 2026-10-16
- PrismML ships Bonsai 2 ternary models: 9B fits in 2GB for local AI — gajesh · 2026-10-16
- Apple MLX Runs 816B Muse Spark Open at 36+ Tokens/s on Four Mac Studios — andrejusb · 2026-10-16
- vLLM Ships Day-0 Support for Meta's 816B Open-Weight Muse Spark Open — woosuk_k · 2026-10-16
- Same 3 Coding Packets, Two Agents: Claude Opus 5.5 Beats Gemini 3.8 Flash on Every One — umeshonai · 2026-10-16