Kimi K3
Moonshot AI · Open weights · Released Jul 16, 2026 · moonshotai/kimi-k3
1.05M contextVisionToolsReasoningStructured outputPrompt cachingOpen weights
Available from 15 providersUSD / MTOK · VERIFIED 12 HOURS AGO
EST. MONTHLY COSTclick a row for detail · click headers to sort
| PROVIDER | INPUT | OUTPUT | CACHE | CONTEXT | QUANT | TTFT | TPS | SMOKE | UPTIME |
|---|---|---|---|---|---|---|---|---|---|
| SRSail Research | $2.60 | $13.00 | $0.29 | 975K | fp4 | 1.5s | 144 tok/s | 1/5 | 98.8% |
| MOMorph | $2.80 | $14.00 | $0.29 | 1.05M | fp4 | 1.7s | 17 tok/s | 0/5 | 98.0% |
 DeepInfra | $2.85 | $14.25 | $0.285 | 1.05M | bf16 | 2s | 25 tok/s | 3/5 | 99.0% |
| DIDigitalOcean | $2.85 | $14.25 | $0.285 | 1.05M | — | 1.2s | 34 tok/s | 3/5 | 100.0% |
| WAWafer | $3.00 | $15.00 | $0.30 | 1.05M | — | 19.6s | — | 0/5 | 83.0% |
 BaseTen | $3.00 | $15.00 | $0.30 | 1.05M | fp8 | 0.61s | 84 tok/s | 1/5 | 99.9% |
 Together | $3.00 | $15.00 | $0.30 | 1.05M | — | 5s | 24 tok/s | — | 98.7% |
 Moonshot AI1ST PARTY | $3.00 | $15.00 | $0.30 | 1.05M | mxfp4 | 3.4s | 86 tok/s | — | 99.6% |
| PHPhala | $3.00 | $15.00 | $0.30 | 1.05M | — | 1.5s | 21 tok/s | — | 99.1% |
 Fireworks | $3.00 | $15.00 | $0.30 | 1.05M | — | 2s | 70 tok/s | — | 99.6% |
| MOModal | $3.00 | $15.00 | $0.30 | 1.05M | mxfp4 | 1.6s | 114 tok/s | — | 99.8% |
| CHChutes | $3.00 | $15.00 | $0.30 | 1.05M | mxfp4 | 1.9s | 33 tok/s | — | 99.1% |
| Fireworks | $3.30 | $16.50 | $0.33 | 1.05M | — | — | — | — | 94.4% |
 Alibaba | $3.45 | $17.25 | $0.345 | 1.05M | — | 1.3s | 18 tok/s | — | 98.0% |
| MOMorph | $6.00 | $22.50 | $0.60 | 1.05M | fp4 | — | — | — | 97.8% |
TTFT/TPS: median (P50) over trailing 72h · chat10k profile (~10K in / 200 out) · measured via OpenRouter routing · SMOKE: of 5 programmatic capability probes passed (JSON schema, instruction following, tool call, long-context, benign compliance; retry-once) · screened via OpenRouter routing · v0.1 · methodology v1.1 · measured 42 hours ago
Pricing variants · USD / MTok in / out
STANDARD
$2.60 / $13.00
per MTok in / out
CACHE READ $0.29
About
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...
Specifications
Context window1,048,576
Max output1,048,576
Modalities intext, image, video
Modalities outtext
LicenseOpen weights
ReleasedJul 16, 2026
TokenizerOther
WeightsHugging Face ↗
Related
Embed badge
[](https://modelindex.ai/models/moonshotai/kimi-k3)Sources
OpenRouter API12 hours ago
AGGREGATORHugging Face12 hours ago
COMMUNITYModelIndex benchmarks42 hours ago
MEASUREDVerified 12 hours ago · list prices, not negotiated rates.