token.app › Moonshot AI › Kimi K2.6
Kimi K2.6 pricing & benchmarks
Kimi K2.6 is available from Moonshot AI at $0.580 per million input tokens and $2.44 per million output tokens ($1.04 blended at 3:1). It accepts up to 262,144 tokens of context. Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration. It handles complex end-to-end coding tasks across Python, Rust, and Go, and...
Pricing
Per million tokens · OpenRouter
Input$0.580
Output$2.44
Blended (3:1)$1.04
Specification
moonshotai/kimi-k2.6
Context window262K
Max output262K
ReleasedApr 2026
LicenceOpen weights
Modalitiesin:text, in:image, out:text
Benchmarks
Independently run — not vendor self-reports. Every score links its source.
| Benchmark | Score | Config | Run | Source |
|---|---|---|---|---|
| GPQA Diamond | 90.8% ±1.7 | — | 2026-05-01 | Independently run · Epoch AI |
| SWE-bench Verified | 76.7% ±1.9 | — | 2026-05-08 | Independently run · Epoch AI |
| FrontierMath T1–3 | 57.2% ±2.9 | — | 2026-06-10 | Independently run · Epoch AI |
| FrontierMath T4 | 25.6% ±7.1 | — | 2026-06-10 | Independently run · Epoch AI |
| SimpleQA Verified | 38.7% ±1.5 | — | 2026-05-01 | Independently run · Epoch AI |
| OTIS Mock AIME | 96.1% ±2.4 | — | 2026-05-02 | Independently run · Epoch AI |
Scores published by Epoch AI, 'AI Benchmarking Hub', used under CC BY 4.0.
Cheaper models that score at least as well on GPQA Diamond
Strictly lower blended price AND an equal-or-higher GPQA Diamond score. One benchmark is one dimension — a model that wins here may still be weaker at your specific task.
| Model | Provider | Blended $/1M | GPQA Diamond | |
|---|---|---|---|---|
| DeepSeek V4 Flash 0731 | DeepSeek | $0.113 (−89%) | 91.0% | Compare → |
| GPT-5.6 Luna | OpenAI | $0.225 (−78%) | 91.6% | Compare → |
| GLM 5.2 | Zhipu AI | $0.387 (−63%) | 91.9% | Compare → |
| DeepSeek V4 Pro | DeepSeek | $0.544 (−48%) | 90.9% | Compare → |
Where to run Kimi K2.6
21 hosts · 1.9× between cheapest and dearest. Blended at 3:1. ⚠ marks an option serving less context than the best available, or re-pricing above a prompt-length threshold. Bracketed labels are the provider's own service tier or region.
| Host | Blended $/1M | In / Out | Context | Weights | Uptime 24h |
|---|---|---|---|---|---|
| Baidu | $1.04 | $0.580 / $2.44 | 262K | fp4 | 99.8% |
| StreamLake | $1.22 | $0.674 / $2.84 | 256K | fp8 | 99.5% |
| Chutes | $1.28 | $0.580 / $3.40 | 262K | int4 | 98.6% |
| Decart | $1.29 | $0.590 / $3.40 | 262K | fp4 | 83.1% |
| Inceptron | $1.30 | $0.600 / $3.41 | 262K | int4 | 99.3% |
| CoreWeave | $1.34 | $0.650 / $3.41 | 262K | fp4 | 98.8% |
| DigitalOcean | $1.37 | $0.760 / $3.20 | 262K | — | 99.1% |
| Crusoe | $1.40 | $0.700 / $3.50 | 262K | bf16 | 99.2% |
| SiliconFlow | $1.43 | $0.770 / $3.40 | 262K | fp8 | 98.8% |
| Parasail | $1.44 | $0.750 / $3.50 | 262K | int4 | 98.7% |
| Venice | $1.44 | $0.750 / $3.50 | 256K | int4 | 76.0% |
| DeepInfra | $1.44 | $0.750 / $3.50 | 262K | fp4 | 99.4% |
| Novita | $1.45 | $0.800 / $3.40 | 262K | — | 99.9% |
| AtlasCloud | $1.71 | $0.950 / $4.00 | 262K | int4 | 98.9% |
| Moonshot AI | $1.71 | $0.950 / $4.00 | 262K | int4 | 100.0% |
| Cloudflare | $1.71 | $0.950 / $4.00 | 262K | — | 99.8% |
| BaseTen | $1.71 | $0.950 / $4.00 | 262K | fp4 | — |
| Fireworks | $1.71 | $0.950 / $4.00 | 262K | — | 0.0% |
| Sail Research | $1.75 | $1.00 / $4.00 | 262K | int4 | 99.6% |
| Phala | $1.97 | $1.09 / $4.60 | 262K | — | 98.6% |
| Together | $2.02 | $1.20 / $4.50 | 262K | — | 67.1% |
Host prices and uptime from OpenRouter, synced 2026-08-08. Lower-precision weights (fp4, fp8) can cost less and score worse than the same model at bf16 — the score above is not host-specific.
Compare Kimi K2.6
FAQ
How much does Kimi K2.6 cost?
Kimi K2.6 costs $0.580 per million input tokens and $2.44 per million output tokens on OpenRouter. At a 3:1 input:output mix that blends to $1.04 per million tokens.
What is Kimi K2.6's context window?
Kimi K2.6 accepts up to 262,144 tokens of context, and can generate up to 262,144 output tokens.
How good is Kimi K2.6 on benchmarks?
Kimi K2.6 scores 90.8% on GPQA Diamond, independently run and published by Epoch AI. 6 benchmark results are listed on this page, each with its run date and source.
Is there a cheaper model as good as Kimi K2.6?
Yes — 4 models in our catalogue cost less per token than Kimi K2.6 and score at least as high on the same benchmark. The cheapest is DeepSeek V4 Flash 0731 at $0.113 per million tokens blended.