Kimi
Moonshot's open-weight models, built for long documents and agent runs.
Kimi is Moonshot AI's model family, open-weight and aimed squarely at long-horizon work: big context windows, strong tool use, and a Code variant tuned for software tasks. K3 is a multimodal reasoning model with a million-token context. If you have a large repository or a long agent run and do not want to pay frontier prices, this is the family to price against. The Kimi chat app is listed separately under Chats.
At a glance — Kimi K2.6
- Cost
- $0.95 in · $4.00 out per 1M tokensMid-range
- Memory
- 262K tokens — about 400 pages
- Reads
- Text · Images
- Can do
- Use tools (agents)
- Structured output
- Thinking mode
- Good at
- Coding: better than 69% · Agentic work: better than 54% · Overall: better than 54%
- Reliability
- 99.99% uptime, last 24h
- Released
- Apr 20, 2026
Advanced details
- API model ID
moonshotai/kimi-k2.6- Max output
- 236K tokens
- Cached input
- $0.16 per 1M
- Tokenizer
- Other
- Coding index
- 61.8
- Agentic work index
- 20.5
- Overall index
- 27.0
Providers
The same model, hosted by different companies. OpenRouter picks one per request and falls back to the next if it fails.
| Provider | Input /1M | Output /1M | Cached /1M | Max output | Uptime 24h |
|---|---|---|---|---|---|
| Baidu · fp4 | $0.434 | $1.828 | $0.073 | 236K | 99.99% |
| Inceptron · int4 | $0.434 | $2.39 | $0.105 | 236K | 98.94% |
| Chutes · int4 | $0.50 | $2.85 | $0.05 | 66K | 99.08% |
| DigitalOcean | $0.57 | $2.40 | $0.114 | 236K | 95.75% |
| Decart · fp4 | $0.587 | $2.47 | $0.099 | 236K | 98.72% |
| StreamLake · fp8 | $0.598 | $2.52 | $0.101 | 230K | 99.40% |
| CoreWeave · fp4 | $0.65 | $3.41 | $0.15 | 236K | 96.87% |
| Crusoe · bf16 | $0.70 | $3.50 | $0.35 | 236K | 67.82% |
| DeepInfra · fp4 | $0.75 | $3.50 | $0.15 | 16K | 98.76% |
| Venice · int4 | $0.75 | $3.50 | $0.16 | 66K | 91.96% |
| Parasail · int4 | $0.75 | $3.50 | $0.16 | 236K | 97.11% |
| SiliconFlow · fp8 | $0.77 | $3.40 | $0.14 | 236K | 99.76% |
| Novita | $0.80 | $3.40 | $0.16 | 236K | 99.73% |
| GMICloud · fp8 | $0.855 | $3.60 | $0.144 | 236K | 97.56% |
| AtlasCloud · int4 | $0.95 | $4.00 | $0.16 | 236K | 98.40% |
| Cloudflare | $0.95 | $4.00 | $0.16 | 236K | 99.93% |
| Moonshot AI · int4 | $0.95 | $4.00 | $0.16 | 236K | 99.88% |
| Fireworks | $0.95 | $4.00 | $0.16 | 236K | 0.00% |
| Phala | $1.09 | $4.60 | $0.37 | 236K | 99.33% |
Supported parameters
- frequency_penalty
- include_reasoning
- logit_bias
- logprobs
- max_tokens
- min_p
- parallel_tool_calls
- presence_penalty
- reasoning
- repetition_penalty
- response_format
- seed
- stop
- structured_outputs
- temperature
- tool_choice
- tools
- top_k
- top_logprobs
- top_p
Live from OpenRouter, refreshed hourly. Scores from Artificial Analysis.
Key info
- Pricing
- Open source
- Category
- Models
- Best for
- Intermediate
- Made by
- Moonshot AI
- Weights
- Open
- Good for
- Long documents and long agent runs