AI - Paski.devKimi K2.6
Moonshot AI llmFollow-up multimodal agentic model improving long-horizon coding, tool use and agentic browsing, with native INT4 quantisation.
overall ⌄
65.4
15 of 93 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.
| benchmark | published | measured |
|---|---|---|
| GPQA | 90.5 | — |
| SWE-bench | 80.2 | — |
| SWE-bench Pro | 58.6 | — |
| HLE | 36.4 | — |
| BrowseComp | 83.2 | — |
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.
| DeepSeek V4-Pro DeepSeek | $0.00130 | overall 68.3 |
Lineage
↓ finetune Kimi K2.7-Code
Metadata
| org | Moonshot AI |
| released | 2026-04-20 |
| params | 1T-A32B |
| license | modified-mit |
| hf id | moonshotai/Kimi-K2.6 |