Kimi K2.6
Kimi K2.6 is an open-weights AI model from Moonshot with an index score of 152 (GPT-4 = 100), rank 64 of 217.
152136–169Index score#64 of 217 models
$0.65/ $3.41Price per million tokensinput / output · via OpenRouter
262KContext windowtokens it reads at once
20 Apr 2026ReleasedReplaced by Kimi K3
Where it sitson the frontier
Gemini 3 Pro reached this score 5 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingevidence
Agentsevidence
Professional workevidence
Novel problemsno evidence yet
Long tasksno evidence yet
Third-party results only, at the model’s best standard setting. Hover a bar for who ran it.
What others sayrank on 11 leaderboards
The Model Indexthird-party tests, our method#64of 217 · 152
LMArena · Mathsmaths questions in chat#35of 346 · 1479LMArena · Codingcoding questions in chat#42of 357 · 1516LMArena · Long promptslong, complex requests#45of 340 · 1478LMArena · Instruction followingfollowing detailed instructions#51of 362 · 1455LMArena · Textoverall chat quality, judged by people#53of 362 · 1461LMArena · Creative writingcreative writing#59of 360 · 1432LMArena · Visionunderstanding images#33of 145 · 1265Epoch Capabilities Indexoverall capability from many benchmarks#61of 231 · 151.6LMArena · WebDevbuilding web apps and front ends#61of 133 · 1510LMArena · Documentsworking with long documents#23of 40 · 1450Best attop three among current models
Where to use itprices checked 18 Nov 2026
API providers 8 of 15 · per 1M tokens, in / out
Run it yourself
Official weightsmoonshotai/Kimi-K2.6↗Host prices from OpenRouter; going direct can differ.
Facts and sources
Released20 Apr 2026
Announcementkimi.com ↗
Release entryKimi K2.6
Scored asKimi K2.6
Output speed–
Hugging Facemoonshotai/Kimi-K2.6 ↗