Kimi K2.6

Moonshot AI llm

Follow-up multimodal agentic model improving long-horizon coding, tool use and agentic browsing, with native INT4 quantisation.

overall ⌄

61.6

31 of 125 ranked

price · $/M tokens

0.95 / 4

in / out, as of 2026-07-25

to run it yourself ⌄

600 GB

minimum · 2.4 TB recommended

Benchmarks

What its maker published, and what anyone else measured. Every figure links the document it came from.

benchmarkpublishedmeasured
GPQA90.5 —
SWE-bench80.2 —
SWE-bench Pro58.6 —
HLE36.4 —
BrowseComp83.2 —

Cheaper, and at least as good

On a 1,000-in, 1,000-out call these cost less and rank no lower — a hard comparison to argue with, and a narrow one. Value covers what it misses.

Claude Haiku 5.5 Anthropic$0.00060overall 65.3
DeepSeek V4-Pro DeepSeek$0.00130overall 65.6
MiniMax M3 MiniMax$0.00150overall 61.8
Qwen3.5 397B A17B Alibaba$0.00420overall 63.3

Lineage

↓ finetune Kimi K2.7-Code
view in full graph →

Metadata

orgMoonshot AI
released2026-04-20
params1T-A32B
licensemodified-mit
hf idmoonshotai/Kimi-K2.6