Kimi K2.6
- Context
- 262.14K
- 262K max output
- Input price
- $0.95
- per 1M tokens · Moonshot AI Platform
- Output price
- $4.00
- per 1M tokens
- Epoch Capabilities Index
- 151
- #38 of 71 listed
- LMArena Text
- 1,461
- #33 of 79 listed
Where to run it
3 offerings · prices as listed| Meter | Tier | Price |
|---|---|---|
| Input | Standard | $0.95 / 1M tokens |
| Output | Standard | $4.00 / 1M tokens |
| Cache read | Standard | $0.16 / 1M tokens |
| Input | Standard | ¥6.50 / 1M tokens |
| Output | Standard | ¥27.00 / 1M tokens |
| Cache read | Standard | ¥1.10 / 1M tokens |
SiliconFlow (opens in a new tab) Pro/moonshotai/Kimi-K2.6 | ||
| Input | Standard | ¥6.50 / 1M tokens |
| Output | Standard | ¥27.00 / 1M tokens |
| Cache read | Standard | ¥1.10 / 1M tokens |
Vendor-reported scores
Published by Moonshot AI, under the conditions it states| Benchmark | Setting | Score |
|---|---|---|
| Humanity's Last Exam | full set, no tools, thinking | 34.7% |
| full set, with search/code/browsing tools, thinking | 54% | |
| BrowseComp | thinking, with tools, discard-all context management | 83.2% |
| Agent Swarm mode | 86.3% | |
| OSWorld-Verified | thinking | 73.1% |
| Terminal-Bench 2.0 | Terminus-2 harness, preserve thinking | 66.7% |
| SWE-bench Pro | in-house SWE-agent-based harness, avg of 10 runs | 58.6% |
| SWE-bench Verified | in-house SWE-agent-based harness, avg of 10 runs | 80.2% |
| LiveCodeBench | v6, thinking | 89.6% |
| AIME 2026 | thinking | 96.4% |
| GPQA Diamond | thinking | 90.5% |
| MMMU-Pro | no tools, thinking, avg@3 | 79.4% |
| with Python tool, thinking, avg@3 | 80.1% |
Cited scores
From third-party leaderboards, not measured by us- Epoch Capabilities Index 151
#38 of 71 listed models · CI 149.2–152.8
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- LMArena Text 1,461
#33 of 79 listed models · CI 1,457–1,465 · 39,971 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made
- Epoch AI · FrontierMath Tiers 1–3 57.2%
#32 of 56 listed models · ±2.94
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- Epoch AI · GPQA Diamond 90.8%
#24 of 64 listed models · ±1.72
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- Epoch AI · SWE-bench Verified 76.7%
#9 of 23 listed models · ±1.92
Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Sep 26, 2026 · changes we made
- LMArena Text · Chinese 1,517
#24 of 79 listed models · CI 1,504–1,530 · 2,343 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made
- LMArena Text · Coding 1,516
#30 of 79 listed models · CI 1,509–1,523 · 11,092 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 25, 2026 · changes we made
- LMArena Vision 1,280
#19 of 49 listed models · CI 1,273–1,287 · 15,347 votes
LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Sep 13, 2026 · changes we made