Kimi K3
Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at navigating large repositories, using tools, debugging, and iterating against images, logs, tests, and runtime feedback. Its architecture uses KDA and Attention Residuals for computational efficiency.
- Organization
- Moonshot AI
- Family
- kimi-k3
- Providers
- 84
- Context
- 1,048,576
- Output limit
- 131,072
- Knowledge
- —
- Release
- 2026-07-16
- Updated
- —
- Weights
- Open
- Input
- text, image/jpeg, image/png, image/gif, image/webp, image/bmp, image/heic, image/heif, video/mp4, video/mpeg, video/mov, video/avi, video/x-flv, video/mpg, video/webm, video/wmv, video/3gpp
- Output
- text
- Capabilities
- —
Providers
Compare every cataloged offer for this model, including its provider, route, token limits, price, and capabilities.
| Provider | Model ID | Context | Output | Input / output · 1M | Reasoning | Tools | Structured | Details |
|---|---|---|---|---|---|---|---|---|
| 302.AI | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Abacus | moonshotai/Kimi-K3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| ai& | moonshotai/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| AIHubMix | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| AI/ML API | moonshot/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| ainetcafe | Kimi-K3 | 262,144 | 32,768 | — / — | — | — | — | View |
| Alibaba | kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| Alibaba (China) | kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| Amazon Bedrock | us.moonshotai.kimi-k3 | 1,048,576 | 128,000 | — / — | — | — | — | View |
| Amazon Bedrock | global.moonshotai.kimi-k3 | 1,048,576 | 128,000 | — / — | — | — | — | View |
| Arcee | moonshotai/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Baseten | moonshotai/Kimi-K3 | 1,048,576 | 262,144 | — / — | — | — | — | View |
| Berget.AI | moonshotai/Kimi-K3 | 327,680 | 32,768 | — / — | — | — | — | View |
| Boundless | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Charm Hyper | kimi-k3 | 1,048,576 | 16,000 | — / — | — | — | — | View |
| Chutes | moonshotai/Kimi-K3-TEE | 1,048,576 | 65,535 | — / — | — | — | — | View |
| ClinePass | cline-pass/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Cloudflare AI Gateway | moonshotai/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| CommandCode | moonshotai/Kimi-K3 | 1,000,000 | 131,072 | — / — | — | — | — | View |
| Cortecs | kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| CrossModel | moonshot/kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| Deep Infra | moonshotai/Kimi-K3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| DevPass (LLM Gateway) | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| DevPass (LLM Gateway) | kimi-k3-fast | 1,040,384 | 131,072 | — / — | — | — | — | View |
| DigitalOcean | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Eden AI | moonshot/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| EmpirioLabs AI | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| engy | kimi-k3 | 1,113,088 | 65,536 | — / — | — | — | — | View |
| Fireworks AI | accounts/fireworks/models/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Fireworks AI | accounts/fireworks/routers/kimi-k3-fast | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Fireworks AI | accounts/fireworks/routers/kimi-k3-spot | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Fireworks AI | accounts/fireworks/routers/kimi-latest | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Fireworks AI | accounts/fireworks/routers/kimi-fast-latest | 1,048,576 | 131,072 | — / — | — | — | — | View |
| FrogBot | kimi-k3 | 1,040,000 | 131,072 | — / — | — | — | — | View |
| GitHub Copilot | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| GreenPT | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Hugging Face | moonshotai/Kimi-K3 | 1,000,000 | 131,072 | — / — | — | — | — | View |
| Impossibl | moonshotai/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Inco | kimi-k3:fast | 1,048,576 | 131,072 | — / — | — | — | — | View |
| IteraCompute | moonshotai/kimi-k3 | 1,048,576 | 999,999 | — / — | — | — | — | View |
| Jalapeno Cloud | Kimi-K3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Kenari | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Kilo Gateway | moonshotai/kimi-k3 | 1,048,576 | 943,718 | — / — | — | — | — | View |
| Kilo Gateway | moonshotai/kimi-k3-fast | 1,000,000 | 1,000,000 | — / — | — | — | — | View |
| Kimi For Coding (kimi.ai) | k3-256k | 262,144 | 131,072 | — / — | — | — | — | View |
| Kimi For Coding (kimi.ai) | k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Kimi For Coding (kimi.com) | k3-256k | 262,144 | 131,072 | — / — | — | — | — | View |
| Kimi For Coding (kimi.com) | k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| LLM Gateway | aws-bedrock/kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| LLM Gateway | together-ai/kimi-k3 | 1,000,000 | 1,000,000 | — / — | — | — | — | View |
| LLM Gateway | novita/kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| LLM Gateway | baidu/kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| LLM Gateway | scx-ai-gp/kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| LLM Gateway | tencent/kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| LLM Gateway | fireworks/kimi-k3 | 1,040,384 | 1,040,384 | — / — | — | — | — | View |
| LLM Gateway | canopywave/kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| LLM Gateway | runpod/kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| LLM Gateway | runware/kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| LLM Gateway | moonshot/kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| LLM Gateway | alibaba/kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| LLM Gateway | fireworks/kimi-k3-fast | 1,040,384 | 1,040,384 | — / — | — | — | — | View |
| Melious | kimi-k3 | 1,000,000 | 131,072 | — / — | — | — | — | View |
| Merge Gateway | moonshot/kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| Moark | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Modal | moonshotai/Kimi-K3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Moonshot AI | kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| Moonshot AI (China) | kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| Morph | morph-kimik3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| Morph | morph-kimik3-fast | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| NanoGPT | TEE/kimi-k3 | 1,048,576 | 65,535 | — / — | — | — | — | View |
| NanoGPT | moonshotai/kimi-k3 | 1,048,576 | 943,718 | — / — | — | — | — | View |
| NEAR AI Cloud | moonshotai/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Nebius Token Factory | moonshotai/Kimi-K3 | 1,024,000 | 8,000 | — / — | — | — | — | View |
| Neon | kimi-k3 | 1,048,576 | 65,536 | — / — | — | — | — | View |
| Neuralwatt | kimi-k3-flex | 1,048,560 | 1,048,560 | — / — | — | — | — | View |
| Neuralwatt | kimi-k3 | 1,048,560 | 1,048,560 | — / — | — | — | — | View |
| Neuralwatt | kimi-k3-fast | 1,048,560 | 1,048,560 | — / — | — | — | — | View |
| Novita AI | moonshotai/kimi-k3 | 1,048,576 | 1,048,576 | — / — | — | — | — | View |
| Nvidia | moonshotai/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Ofox | moonshotai/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Ollama Cloud | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| OpenCode | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| OpenCode Go | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| AkashMLvia OpenRouter | moonshotai/kimi-k3 fp4 | 1,048,576 | 943,718 | $1.20 / $14.00 | Yes | Yes | Yes | View |
| Alibabavia OpenRouter | moonshotai/kimi-k3 unknown | 1,048,576 | 131,072 | $3.45 / $17.25 | Yes | Yes | Yes | View |
| Amazon Bedrockvia OpenRouter | moonshotai/kimi-k3 unknown | 1,048,576 | 128,000 | $3.00 / $15.00 | Yes | Yes | No | View |
| BaseTenvia OpenRouter | moonshotai/kimi-k3 fp8 | 1,048,576 | 262,144 | $3.00 / $15.00 | Yes | Yes | No | View |
| Chutesvia OpenRouter | moonshotai/kimi-k3 mxfp4 | 1,048,576 | 65,535 | $3.00 / $15.00 | Yes | No | Yes | View |
| Decartvia OpenRouter | moonshotai/kimi-k3 mxfp4 | 1,048,576 | 943,718 | $1.98 / $9.90 | Yes | Yes | Yes | View |
| DeepInfravia OpenRouter | moonshotai/kimi-k3 mxfp4 | 1,048,576 | 943,718 | $2.85 / $14.25 | Yes | Yes | Yes | View |
| DigitalOceanvia OpenRouter | moonshotai/kimi-k3 unknown | 1,048,576 | 943,718 | $2.55 / $12.95 | Yes | Yes | Yes | View |
| Fireworksvia OpenRouter | moonshotai/kimi-k3 unknown | 1,048,576 | 943,718 | $4.50 / $22.50 | Yes | Yes | Yes | View |
| Fireworksvia OpenRouter | moonshotai/kimi-k3 unknown | 1,048,576 | 943,718 | $4.50 / $22.50 | Yes | Yes | Yes | View |
| Fireworksvia OpenRouter | moonshotai/kimi-k3 unknown | 1,048,576 | 943,718 | $3.00 / $15.00 | Yes | Yes | Yes | View |
| InferenceNetvia OpenRouter | moonshotai/kimi-k3 fp4 | 250,000 | 225,000 | $2.99 / $15.00 | Yes | Yes | Yes | View |
| InferenceNetvia OpenRouter | moonshotai/kimi-k3 fp4 | 1,048,576 | 943,718 | $0.64 / $13.50 | Yes | Yes | Yes | View |
| Makoravia OpenRouter | moonshotai/kimi-k3 fp4 | 1,048,576 | 943,718 | $1.53 / $12.75 | Yes | Yes | Yes | View |
| Modalvia OpenRouter | moonshotai/kimi-k3 mxfp4 | 1,048,576 | 943,718 | $3.00 / $15.00 | Yes | Yes | Yes | View |
| Moonshot AIvia OpenRouter | moonshotai/kimi-k3 mxfp4 | 1,048,576 | 943,718 | $3.00 / $15.00 | Yes | Yes | Yes | View |
| Morphvia OpenRouter | moonshotai/kimi-k3 fp8 | 1,048,576 | 943,718 | $0.339 / $14.90 | Yes | Yes | Yes | View |
| Parasailvia OpenRouter | moonshotai/kimi-k3 fp4 | 1,048,576 | 943,718 | $2.60 / $13.00 | Yes | Yes | Yes | View |
| Parasailvia OpenRouter | moonshotai/kimi-k3 fp4 | 1,000,000 | 900,000 | $3.00 / $15.00 | Yes | No | Yes | View |
| Phalavia OpenRouter | moonshotai/kimi-k3 unknown | 1,048,576 | 943,718 | $2.55 / $12.75 | Yes | Yes | Yes | View |
| Relacevia OpenRouter | moonshotai/kimi-k3 fp4 | 1,048,576 | 943,718 | $2.00 / $14.00 | Yes | Yes | No | View |
| Sail Researchvia OpenRouter | moonshotai/kimi-k3 fp4 | 1,048,576 | 943,718 | $0.84 / $13.50 | Yes | Yes | Yes | View |
| Togethervia OpenRouter | moonshotai/kimi-k3 unknown | 1,048,576 | 943,718 | $2.70 / $13.50 | Yes | Yes | Yes | View |
| Wafervia OpenRouter | moonshotai/kimi-k3 unknown | 1,048,576 | 943,718 | $0.33 / $14.89 | Yes | Yes | Yes | View |
| Wafervia OpenRouter | moonshotai/kimi-k3 unknown | 1,048,576 | 943,718 | $2.80 / $14.00 | Yes | Yes | Yes | View |
| Opper | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| OrcaRouter | kimi/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Perplexity Agent | perplexity/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Perplexity Agent | moonshot-ai/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Pioneer | moonshotai/Kimi-K3 | 1,000,000 | 131,072 | — / — | — | — | — | View |
| Pioneer | moonshotai/Kimi-K3-Fast | 1,000,000 | 131,072 | — / — | — | — | — | View |
| Prime Inference | moonshotai/kimi-k3 | 1,048,576 | 943,718 | — / — | — | — | — | View |
| Requesty | kimi-k3@eu | 1,048,576 | 262,144 | — / — | — | — | — | View |
| Requesty | kimi-k3 | 1,048,576 | 262,144 | — / — | — | — | — | View |
| SCNet Token Plan | Kimi-K3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| SenseNova (China) | kimi-k3 | 1,048,576 | 65,536 | — / — | — | — | — | View |
| SiliconFlow | moonshotai/Kimi-K3 | 1,048,576 | 262,000 | — / — | — | — | — | View |
| Standard Compute | moonshotai/kimi-k3 | 1,048,576 | 50,000 | — / — | — | — | — | View |
| Synthetic | hf:moonshotai/Kimi-K3 | 524,288 | 65,536 | — / — | — | — | — | View |
| Tempr Gateway | moonshot/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| TensorX | moonshotai/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Together AI | moonshotai/Kimi-K3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| TokenGo | moonshotai/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Umans AI | umans-kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Umans AI Coding Plan | umans-kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Vancine | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Venice AI | kimi-k3 | 1,000,000 | 131,072 | — / — | — | — | — | View |
| Venice AI | kimi-k3-fast-api | 1,000,000 | 131,072 | — / — | — | — | — | View |
| Vercel AI Gateway | moonshotai/kimi-k3 | 1,000,000 | 131,072 | — / — | — | — | — | View |
| Vercel AI Gateway | moonshotai/kimi-k3-fast | 1,000,000 | 1,000,000 | — / — | — | — | — | View |
| Vivgrid | kimi-k3 | 1,000,000 | 131,072 | — / — | — | — | — | View |
| Volcengine Ark Coding Plan | kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Wallaby | moonshotai/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
| Wally | kimi-k3 | 1,048,576 | 65,536 | — / — | — | — | — | View |
| ZenMux | moonshotai/kimi-k3 | 1,048,576 | 131,072 | — / — | — | — | — | View |
Prices and limits apply to each offer. A dash means the value is not provided. OpenRouter offers show the hosting provider and routing channel separately.
Pricing
Compare token rates and cache charges across available routes and hosting configurations.
Lowest listed input and output prices across the offers above; they may belong to different providers. These are base rates, before conditional pricing, discounts or additional fees.
moonshotai/kimi-k3Observed - Input
- $0.64
- Output
- $13.50
- Cache read
- $0.28
Hosting prices through OpenRouter All endpoints and cache rates
| Endpoint | Input / 1M | Output / 1M | Cache read / 1M | Cache write / 1M |
|---|---|---|---|---|
| AkashMLakashml/fp4 | $1.20 | $14.00 | $1.20 | — |
| Alibabaalibaba | $3.45 | $17.25 | $0.345 | — |
| Amazon Bedrockamazon-bedrock/us-east-2 | $3.00 | $15.00 | $0.30 | $3.75 |
| BaseTenbaseten/fp8 | $3.00 | $15.00 | $0.30 | — |
| Chuteschutes/mxfp4 | $3.00 | $15.00 | $0.30 | — |
| Decartdecart/mxfp4 | $1.98 | $9.90 | $0.198 | — |
| DeepInfradeepinfra/mxfp4 | $2.85 | $14.25 | $0.285 | — |
| DigitalOceandigitalocean | $2.55 | $12.95 | $0.255 | — |
| Fireworksfireworks/us | $4.50 | $22.50 | $0.45 | — |
| Fireworksfireworks/fast | $4.50 | $22.50 | $0.45 | — |
| Fireworksfireworks | $3.00 | $15.00 | $0.30 | — |
| InferenceNetinference-net/fast | $2.99 | $15.00 | $0.45 | — |
| InferenceNetinference-net/fp4 | $0.64 | $13.50 | $0.28 | — |
| Makoramakora/fp4 | $1.53 | $12.75 | $0.204 | — |
| Modalmodal/mxfp4 | $3.00 | $15.00 | $0.30 | — |
| Moonshot AImoonshotai/mxfp4 | $3.00 | $15.00 | $0.30 | — |
| Morphmorph/fp8 | $0.339 | $14.90 | $0.29 | — |
| Parasailparasail/fp4 | $2.60 | $13.00 | $0.26 | — |
| Parasailparasail/fast | $3.00 | $15.00 | $0.30 | — |
| Phalaphala | $2.55 | $12.75 | $0.255 | — |
| Relacerelace/fp4 | $2.00 | $14.00 | $0.30 | — |
| Sail Researchsail-research/fp4 | $0.84 | $13.50 | $0.30 | — |
| Togethertogether | $2.70 | $13.50 | $0.27 | — |
| Waferwafer | $0.33 | $14.89 | $0.30 | — |
| Waferwafer/us | $2.80 | $14.00 | $0.30 | — |
Each row is a hosting configuration. Open an offer in Providers for its full pricing conditions and additional fees. Endpoint observations may differ from the routing catalog's base price.
Performance
Review reported latency and output-throughput percentiles from the latest observed rolling window.
Time to first token · milliseconds
All latency percentiles 25 endpoints
| Endpoint | P50 | P75 | P90 | P99 |
|---|---|---|---|---|
| AkashMLakashml/fp4 | 1,202.5 ms | 1,386.75 ms | 1,525.9 ms | 2,526.97 ms |
| Alibabaalibaba | 1,811 ms | 2,335 ms | 3,124.4 ms | 7,935.19 ms |
| Amazon Bedrockamazon-bedrock/us-east-2 | 1,021 ms | 1,249.5 ms | 1,656.3 ms | 3,976.32 ms |
| BaseTenbaseten/fp8 | 4,510 ms | 6,458.75 ms | 8,998.5 ms | 19,793.8 ms |
| Chuteschutes/mxfp4 | 2,011 ms | 2,396 ms | 4,184 ms | 8,439.96 ms |
| Decartdecart/mxfp4 | 2,501 ms | 10,002.5 ms | 16,042.3 ms | 23,740.59 ms |
| DeepInfradeepinfra/mxfp4 | 1,858.5 ms | 2,888.75 ms | 5,489.7 ms | 16,009.35 ms |
| DigitalOceandigitalocean | 2,364 ms | 3,200 ms | 4,129.6 ms | 11,511.34 ms |
| Fireworksfireworks/us | 1,212.5 ms | 1,582.5 ms | 2,292.5 ms | 5,933.5 ms |
| Fireworksfireworks/fast | 1,114 ms | 2,330.5 ms | 4,301 ms | 13,475.39 ms |
| Fireworksfireworks | 1,019 ms | 1,612 ms | 3,055 ms | 5,197.26 ms |
| InferenceNetinference-net/fast | 833 ms | 1,256 ms | 1,798 ms | 4,535.28 ms |
| InferenceNetinference-net/fp4 | 893 ms | 1,422 ms | 2,237.8 ms | 8,562.42 ms |
| Makoramakora/fp4 | 1,820.5 ms | 3,070.75 ms | 6,500.1 ms | 48,950.79 ms |
| Modalmodal/mxfp4 | 1,241 ms | 2,117 ms | 3,487.2 ms | 13,218.04 ms |
| Moonshot AImoonshotai/mxfp4 | 4,182 ms | 5,319 ms | 7,266.9 ms | 33,236.95 ms |
| Morphmorph/fp8 | 1,267 ms | 2,277.5 ms | 3,204.8 ms | 6,445.56 ms |
| Parasailparasail/fp4 | 1,232.5 ms | 1,934.75 ms | 3,191.8 ms | 11,077.1 ms |
| Parasailparasail/fast | 974 ms | 1,433 ms | 1,788.4 ms | 9,759.36 ms |
| Phalaphala | 1,450 ms | 2,423 ms | 4,108.6 ms | 12,515.5 ms |
| Relacerelace/fp4 | 1,518 ms | 2,489.5 ms | 6,403.2 ms | 29,965.6 ms |
| Sail Researchsail-research/fp4 | 2,518 ms | 3,812 ms | 7,600 ms | 19,890 ms |
| Togethertogether | 1,072 ms | 1,459 ms | 2,015 ms | 5,948.12 ms |
| Waferwafer | 1,015 ms | 1,749.75 ms | 3,886 ms | 21,463.49 ms |
| Waferwafer/us | 738 ms | 1,380 ms | 2,609 ms | 13,866.2 ms |
Output throughput · tokens per second
All throughput percentiles 25 endpoints
| Endpoint | P50 | P75 | P90 | P99 |
|---|---|---|---|---|
| AkashMLakashml/fp4 | 17 t/s | 20 t/s | 24 t/s | 29.87 t/s |
| Alibabaalibaba | 38 t/s | 44 t/s | 50 t/s | 68 t/s |
| Amazon Bedrockamazon-bedrock/us-east-2 | 72 t/s | 88 t/s | 101 t/s | 141 t/s |
| BaseTenbaseten/fp8 | 48 t/s | 71 t/s | 91 t/s | 120.08 t/s |
| Chuteschutes/mxfp4 | 26 t/s | 32 t/s | 33 t/s | 34 t/s |
| Decartdecart/mxfp4 | 22 t/s | 32 t/s | 45 t/s | 81.54 t/s |
| DeepInfradeepinfra/mxfp4 | 40 t/s | 48 t/s | 58 t/s | 79 t/s |
| DigitalOceandigitalocean | 34 t/s | 45 t/s | 53 t/s | 62 t/s |
| Fireworksfireworks/us | 74 t/s | 89 t/s | 103.2 t/s | 114 t/s |
| Fireworksfireworks/fast | 67 t/s | 93.5 t/s | 118.6 t/s | 147.72 t/s |
| Fireworksfireworks | 39 t/s | 43 t/s | 46 t/s | 53 t/s |
| InferenceNetinference-net/fast | 63 t/s | 88 t/s | 112 t/s | 163.44 t/s |
| InferenceNetinference-net/fp4 | 85 t/s | 114 t/s | 144 t/s | 211 t/s |
| Makoramakora/fp4 | 52 t/s | 67 t/s | 80.9 t/s | 96.79 t/s |
| Modalmodal/mxfp4 | 79 t/s | 103 t/s | 131 t/s | 223 t/s |
| Moonshot AImoonshotai/mxfp4 | 24 t/s | 32 t/s | 37 t/s | 44 t/s |
| Morphmorph/fp8 | 50 t/s | 64 t/s | 81 t/s | 113.6 t/s |
| Parasailparasail/fp4 | 65 t/s | 92 t/s | 126 t/s | 188 t/s |
| Parasailparasail/fast | 65 t/s | 86 t/s | 130 t/s | 162.02 t/s |
| Phalaphala | 49 t/s | 68 t/s | 93 t/s | 149 t/s |
| Relacerelace/fp4 | 52 t/s | 79 t/s | 114 t/s | 182.33 t/s |
| Sail Researchsail-research/fp4 | 37 t/s | 62 t/s | 85 t/s | 134 t/s |
| Togethertogether | 48 t/s | 56 t/s | 64 t/s | 81 t/s |
| Waferwafer | 47 t/s | 59 t/s | 71 t/s | 95 t/s |
| Waferwafer/us | 50 t/s | 63 t/s | 78 t/s | 114 t/s |
Percentiles describe the observed request distribution. These are snapshots of a rolling window, not a historical time series.
Uptime
Compare successful request rates across the latest 5-minute, 30-minute, and 24-hour observations.
Reported by OpenRouter at the time of collection; rate-limited requests are excluded. A dash means no measurement was reported. This is not a live availability check.
Benchmarks
Review attributed evaluation results on each source's original scale.
moonshotai/kimi-k3Observed | Design Arena | Category | Elo | Win rate | Rank |
|---|---|---|---|---|
| agents | agenticgamedev | 1,252 | 55% | 3 |
| agents | androidnative | 1,262 | 55.4% | 5 |
| agents | fullstack | 1,315 | 61.8% | 2 |
| agents | godotgamedev | 1,223 | 48.2% | 14 |
| agents | htmlslides | 1,256 | 59.6% | 2 |
| agents | mobileapps | 1,263 | 56.9% | 4 |
| agents | python-pptxslides | 1,273 | 59.2% | 4 |
| agents | webapps | 1,285 | 59.1% | 4 |
| models | 3d | 1,395 | 67.6% | 4 |
| models | codecategories | 1,367 | 63% | 2 |
| models | dataviz | 1,353 | 63.2% | 4 |
| models | gamedev | 1,367 | 61% | 5 |
| models | svg | 1,305 | 59.8% | 4 |
| models | uicomponent | 1,355 | 61.3% | 4 |
| models | website | 1,341 | 59.8% | 2 |
Sources: Artificial Analysis and Design Arena, via OpenRouter. Scores retain their original scales and are not OpenAI Suite ratings.
Detailed evaluations
| Evaluation | Configuration | Source | Results | Details |
|---|---|---|---|---|
| models · gamedev | Kimi K3moonshotai/kimi-k3-20260715 | Design Arena | 1,353 Elo | Details |
| models · svg | Kimi K3moonshotai/kimi-k3-20260715 | Design Arena | 1,295 Elo | Details |
| models · 3d | Kimi K3moonshotai/kimi-k3-20260715 | Design Arena | 1,365 Elo | Details |
| models · dataviz | Kimi K3moonshotai/kimi-k3-20260715 | Design Arena | 1,350 Elo | Details |
| models · uicomponent | Kimi K3moonshotai/kimi-k3-20260715 | Design Arena | 1,341 Elo | Details |
| models · codecategories | Kimi K3moonshotai/kimi-k3-20260715 | Design Arena | 1,351 Elo | Details |
| Composite indices | Kimi K3 (Low)moonshotai/kimi-k3-20260715 | Artificial Analysis | Intelligence 17.4 · Coding 72 · Agentic 29.3 | Details |
| tau bench verified airline | MoonshotAI: Kimi K3moonshotai/kimi-k3-20260715 | OpenRouter | 73.852% | Details |
| gpqa diamond | MoonshotAI: Kimi K3moonshotai/kimi-k3-20260715 | OpenRouter | 92.02% | Details |
| models · website | Kimi K3moonshotai/kimi-k3-20260715 | Design Arena | 1,336 Elo | Details |
Sources: Artificial Analysis, Design Arena and OpenRouter. Evaluation configurations and scales differ; results are not interchangeable. Updated Nov 14, 2026.
Apps & session costs
Compare observed median session costs for applications using this model, grouped by conversation length.
| Application | Model route | Turns | Median cost / session | Details |
|---|---|---|---|---|
| Codex | moonshotai/kimi-k3-20260715 | 1 turn | $0.01129 | Details |
| Hermes Agent | moonshotai/kimi-k3-20260715 | 2–9 turns | $0.059208 | Details |
| Claude Code | moonshotai/kimi-k3-20260715 | 50+ turns | $3.86745 | Details |
| Claude Code | moonshotai/kimi-k3-20260715 | 10–49 turns | $0.563408 | Details |
| Kilo Code | moonshotai/kimi-k3-20260715 | 50+ turns | $4.563072 | Details |
| Claude Code | moonshotai/kimi-k3-20260715 | 2–9 turns | $0.080102 | Details |
| Kilo Code | moonshotai/kimi-k3-20260715 | 10–49 turns | $0.799507 | Details |
| Hermes Agent | moonshotai/kimi-k3-20260715 | 1 turn | $0.010316 | Details |
| Claude Code | moonshotai/kimi-k3-20260715 | 1 turn | $0.015466 | Details |
| Hermes Agent | moonshotai/kimi-k3-20260715 | 50+ turns | $2.735942 | Details |
| Kilo Code | moonshotai/kimi-k3-20260715 | 2–9 turns | $0.161886 | Details |
| Hermes Agent | moonshotai/kimi-k3-20260715 | 10–49 turns | $0.473754 | Details |
Source: OpenRouter, as of Nov 9, 2026. Published under CC BY 4.0. These are observed costs per session, not token prices or a forecast for your workload. Apps are compared separately.
Activity
Daily prompt and completion tokens reported for this model's routes in OpenRouter's top 50. Missing days have no published value.
141,264,448,246 tokens
183,651,256,582 tokens
177,612,676,416 tokens
205,104,101,779 tokens
210,125,954,459 tokens
213,772,516,280 tokens
172,118,663,922 tokens
159,871,280,445 tokens
225,409,049,776 tokens
236,868,165,710 tokens
237,565,016,933 tokens
247,725,381,861 tokens
225,998,184,201 tokens
170,222,788,817 tokens
230,510,697,288 tokens
240,776,882,141 tokens
304,647,563,810 tokens
295,747,954,459 tokens
341,422,677,144 tokens
321,216,396,862 tokens
296,311,315,639 tokens
179,719,297,597 tokens
223,578,066,849 tokens
258,659,434,488 tokens
254,307,352,004 tokens
197,083,498,590 tokens
203,231,210,868 tokens
167,450,711,446 tokens
167,235,049,261 tokens
203,962,504,573 tokens
204,411,193,019 tokens
226,330,895,059 tokens
222,257,940,094 tokens
230,212,969,990 tokens
192,220,446,293 tokens
173,315,800,865 tokens
217,507,005,740 tokens
192,626,651,557 tokens
188,682,741,083 tokens
195,628,585,826 tokens
211,331,435,630 tokens
185,196,307,594 tokens
203,339,351,480 tokens
228,542,070,956 tokens
275,669,934,181 tokens
263,544,817,299 tokens
203,902,650,973 tokens
211,604,184,550 tokens
210,928,154,287 tokens
225,109,546,128 tokens
299,471,440,010 tokens
281,721,532,479 tokens
285,885,043,308 tokens
328,787,105,885 tokens
338,553,317,971 tokens
307,169,340,883 tokens
308,653,041,997 tokens
452,261,046,008 tokens
497,238,529,260 tokens
393,654,343,806 tokens
236,085,134,296 tokens
203,418,976,604 tokens
179,817,658,144 tokens
250,191,500,936 tokens
283,384,167,184 tokens
278,502,113,230 tokens
285,359,271,054 tokens
413,236,310,721 tokens
341,783,929,722 tokens
189,036,027,587 tokens
198,369,617,568 tokens
196,571,686,110 tokens
300,058,909,714 tokens
316,379,304,983 tokens
273,217,244,405 tokens
228,548,221,968 tokens
189,810,173,007 tokens
201,565,909,640 tokens
260,952,704,993 tokens
348,625,452,654 tokens
669,388,804,332 tokens
211,045,320,429 tokens
186,743,007,506 tokens
107,352,632,470 tokens
119,955,205,391 tokens
148,943,749,487 tokens
185,145,676,564 tokens
184,697,832,117 tokens
199,817,487,718 tokens
198,620,426,174 tokens
Daily observations · 90 days
| Date (UTC) | Reported tokens | Reported routes |
|---|---|---|
| Nov 13, 2026 | 198,620,426,174 | 1 |
| Nov 12, 2026 | 199,817,487,718 | 1 |
| Nov 11, 2026 | 184,697,832,117 | 1 |
| Nov 10, 2026 | 185,145,676,564 | 1 |
| Nov 9, 2026 | 148,943,749,487 | 1 |
| Nov 8, 2026 | 119,955,205,391 | 1 |
| Nov 7, 2026 | 107,352,632,470 | 1 |
| Nov 6, 2026 | 186,743,007,506 | 1 |
| Nov 5, 2026 | 211,045,320,429 | 1 |
| Nov 4, 2026 | 669,388,804,332 | 1 |
| Nov 3, 2026 | 348,625,452,654 | 1 |
| Nov 2, 2026 | 260,952,704,993 | 1 |
| Nov 1, 2026 | 201,565,909,640 | 1 |
| Oct 31, 2026 | 189,810,173,007 | 1 |
| Oct 30, 2026 | 228,548,221,968 | 1 |
| Oct 29, 2026 | 273,217,244,405 | 1 |
| Oct 28, 2026 | 316,379,304,983 | 1 |
| Oct 27, 2026 | 300,058,909,714 | 1 |
| Oct 26, 2026 | 196,571,686,110 | 1 |
| Oct 25, 2026 | 198,369,617,568 | 1 |
| Oct 24, 2026 | 189,036,027,587 | 1 |
| Oct 23, 2026 | 341,783,929,722 | 1 |
| Oct 22, 2026 | 413,236,310,721 | 1 |
| Oct 21, 2026 | 285,359,271,054 | 1 |
| Oct 20, 2026 | 278,502,113,230 | 1 |
| Oct 19, 2026 | 283,384,167,184 | 1 |
| Oct 18, 2026 | 250,191,500,936 | 1 |
| Oct 17, 2026 | 179,817,658,144 | 1 |
| Oct 16, 2026 | 203,418,976,604 | 1 |
| Oct 15, 2026 | 236,085,134,296 | 1 |
| Oct 14, 2026 | 393,654,343,806 | 1 |
| Oct 13, 2026 | 497,238,529,260 | 1 |
| Oct 12, 2026 | 452,261,046,008 | 1 |
| Oct 11, 2026 | 308,653,041,997 | 1 |
| Oct 10, 2026 | 307,169,340,883 | 1 |
| Oct 9, 2026 | 338,553,317,971 | 1 |
| Oct 8, 2026 | 328,787,105,885 | 1 |
| Oct 7, 2026 | 285,885,043,308 | 1 |
| Oct 6, 2026 | 281,721,532,479 | 1 |
| Oct 5, 2026 | 299,471,440,010 | 1 |
| Oct 4, 2026 | 225,109,546,128 | 1 |
| Oct 3, 2026 | 210,928,154,287 | 1 |
| Oct 2, 2026 | 211,604,184,550 | 1 |
| Oct 1, 2026 | 203,902,650,973 | 1 |
| Sep 30, 2026 | 263,544,817,299 | 1 |
| Sep 29, 2026 | 275,669,934,181 | 1 |
| Sep 28, 2026 | 228,542,070,956 | 1 |
| Sep 27, 2026 | 203,339,351,480 | 1 |
| Sep 26, 2026 | 185,196,307,594 | 1 |
| Sep 25, 2026 | 211,331,435,630 | 1 |
| Sep 24, 2026 | 195,628,585,826 | 1 |
| Sep 23, 2026 | 188,682,741,083 | 1 |
| Sep 22, 2026 | 192,626,651,557 | 1 |
| Sep 21, 2026 | 217,507,005,740 | 1 |
| Sep 20, 2026 | 173,315,800,865 | 1 |
| Sep 19, 2026 | 192,220,446,293 | 1 |
| Sep 18, 2026 | 230,212,969,990 | 1 |
| Sep 17, 2026 | 222,257,940,094 | 1 |
| Sep 16, 2026 | 226,330,895,059 | 1 |
| Sep 15, 2026 | 204,411,193,019 | 1 |
| Sep 14, 2026 | 203,962,504,573 | 1 |
| Sep 13, 2026 | 167,235,049,261 | 1 |
| Sep 12, 2026 | 167,450,711,446 | 1 |
| Sep 11, 2026 | 203,231,210,868 | 1 |
| Sep 10, 2026 | 197,083,498,590 | 1 |
| Sep 9, 2026 | 254,307,352,004 | 1 |
| Sep 8, 2026 | 258,659,434,488 | 1 |
| Sep 7, 2026 | 223,578,066,849 | 1 |
| Sep 6, 2026 | 179,719,297,597 | 1 |
| Sep 5, 2026 | 296,311,315,639 | 1 |
| Sep 4, 2026 | 321,216,396,862 | 1 |
| Sep 3, 2026 | 341,422,677,144 | 1 |
| Sep 2, 2026 | 295,747,954,459 | 1 |
| Sep 1, 2026 | 304,647,563,810 | 1 |
| Aug 31, 2026 | 240,776,882,141 | 1 |
| Aug 30, 2026 | 230,510,697,288 | 1 |
| Aug 29, 2026 | 170,222,788,817 | 1 |
| Aug 28, 2026 | 225,998,184,201 | 1 |
| Aug 27, 2026 | 247,725,381,861 | 1 |
| Aug 26, 2026 | 237,565,016,933 | 1 |
| Aug 25, 2026 | 236,868,165,710 | 1 |
| Aug 24, 2026 | 225,409,049,776 | 1 |
| Aug 23, 2026 | 159,871,280,445 | 1 |
| Aug 22, 2026 | 172,118,663,922 | 1 |
| Aug 21, 2026 | 213,772,516,280 | 1 |
| Aug 20, 2026 | 210,125,954,459 | 1 |
| Aug 19, 2026 | 205,104,101,779 | 1 |
| Aug 18, 2026 | 177,612,676,416 | 1 |
| Aug 17, 2026 | 183,651,256,582 | 1 |
| Aug 16, 2026 | 141,264,448,246 | 1 |
Source: OpenRouter (openrouter.ai/rankings), as of Nov 14, 2026. CC BY 4.0. Totals include only individually published routes. Tokenizers differ by provider; the aggregated “other” category is never assigned to a model.
Technical details & API
Explore specifications, supported parameters, reasoning controls, and API identifiers for each OpenRouter route.
moonshotai/kimi-k3Observed - Context
- 1,048,576
- Maximum output
- 943,718
- Tokenizer
- Other
- Input
- text, image, video
- Output
- text
- Moderation
- No
- Added to OpenRouter
- 2026-07-16
https://openrouter.ai/api/v1moonshotai/kimi-k3moonshotai/kimi-k3-20260715moonshotai/Kimi-K3Reasoning
- Required
- No
- Enabled by default
- Yes
- Default effort
- max
- Supported efforts
- max, high, low
Supported parameters
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
Default parameters
- top p
- 0.95
Frequently asked questions
Answers based on the model metadata and serving offers currently in the catalog.
How much does Kimi K3 cost?
Listed rates start at $0.33 per million input tokens and $9.90 per million output tokens. Prices vary by provider and configuration; see Pricing for details.
Which providers offer Kimi K3?
There are 138 cataloged offers across 84 providers. Compare model IDs, prices and limits in Providers.
What is the context window?
The cataloged model context is 1,048,576 tokens. Each serving endpoint may apply a different limit.
Does it support tools and structured output?
Tool calling: Not reported. Structured output: Not reported. Support can vary by endpoint.
Catalog history · 3 recorded revisions
Metadata observations since this model was first cataloged. These are not model release versions.
- ·
a16d3344dbd8 - ·
0a125e654385 - ·
c7865f02a448