Kimi K3
Moonshot AI · Open weights · Released Jul 16, 2026 · moonshotai/kimi-k3
Best $0.68 in / $9.75 out via Wafer
1.05M contextVisionToolsReasoningStructured outputPrompt cachingOpen weights
Available from 26 providersUSD / MTOK · PRICES 23 MINUTES AGO · SPEED 24 HOURS AGO
EST. MONTHLY COSTclick a row for detail · click headers to sort
| PROVIDER | INPUT | OUTPUT | CACHE | CONTEXT | QUANT | TTFT | TPS | SMOKE | UPTIME |
|---|---|---|---|---|---|---|---|---|---|
 Wafer | $0.68 | $15.00 | $0.45 | 1.05M | — | 0.87s | 69 tok/s | 0/5 | 100.0% |
 Relace | $0.71 | $13.00 | $0.45 | 1.05M | fp4 | 1.1s | 150 tok/s | — | 100.0% |
 InferenceNet | $0.72 | $15.00 | $0.43 | 1.05M | fp4 | ✕ | ✕ | — | 100.0% |
 Morph | $0.827 | $12.32 | $0.278 | 1.05M | fp8 | 1.8s | 33 tok/s | 1/5 | 99.9% |
| SRSail Research | $0.84 | $13.50 | $0.30 | 1.05M | fp4 | ✕ | ✕ | 2/5 | 99.9% |
 AkashML | $1.20 | $14.00 | $1.20 | 1.05M | fp4 | ✕ | ✕ | — | 100.0% |
| MAMakora | $1.53 | $12.75 | $0.204 | 1.05M | — | 1.2s | 56 tok/s | — | 96.8% |
 CheaperInferenceDIRECTMARKETPLACE | $1.95 | $9.75 | — | — | — | ✕ | ✕ | — | — |
 Phala | $1.95 | $9.75 | $0.195 | 1.05M | — | 2.8s | 74 tok/s | — | 100.0% |
| CheaperInferenceDIRECTMARKETPLACE | $2.25 | $11.25 | — | — | — | — | — | — | — |
 Decart | $2.52 | $12.60 | $0.252 | 1.05M | mxfp4 | ✕ | ✕ | — | 96.7% |
 DigitalOcean | $2.55 | $12.95 | $0.255 | 1.05M | — | 1.3s | 38 tok/s | 4/5 | 100.0% |
 Parasail | $2.60 | $13.00 | $0.26 | 1.05M | fp4 | 0.91s | 235 tok/s | — | 99.9% |
 Together | $2.70 | $13.50 | $0.27 | 1.05M | — | ✕ | ✕ | — | 85.9% |
| Wafer | $2.80 | $14.00 | $0.30 | 1.05M | — | — | — | — | 99.5% |
 DeepInfra | $2.85 | $14.25 | $0.285 | 1.05M | mxfp4 | 1.3s | 74 tok/s | 4/5 | 99.1% |
| InferenceNetPRIORITY | $2.99 | $15.00 | $0.45 | 250K | fp4 | — | — | — | 100.0% |
 Amazon Bedrock | $3.00 | $15.00 | $0.30 | 1.05M | — | ✕ | ✕ | — | 99.9% |
 BaseTen | $3.00 | $15.00 | $0.30 | 1.05M | fp8 | ✕ | ✕ | 1/5 | 100.0% |
 Chutes | $3.00 | $15.00 | $0.30 | 1.05M | mxfp4 | ✕ | ✕ | — | 100.0% |
 Fireworks | $3.00 | $15.00 | $0.30 | 1.05M | — | ✕ | ✕ | — | 99.9% |
 Modal | $3.00 | $15.00 | $0.30 | 1.05M | mxfp4 | ✕ | ✕ | — | 100.0% |
 Moonshot AI1ST PARTY | $3.00 | $15.00 | $0.30 | 1.05M | mxfp4 | ✕ | ✕ | — | 100.0% |
 Alibaba | $3.45 | $17.25 | $0.345 | 1.05M | — | 1.6s | 36 tok/s | — | 95.7% |
| Fireworks | $4.50 | $22.50 | $0.45 | 1.05M | — | — | — | — | 100.0% |
| FireworksPRIORITY | $4.50 | $22.50 | $0.45 | 1.05M | — | — | — | — | 99.1% |
TTFT/TPS: median (P50) over trailing 504h · chat10k profile (~10K in / 200 out) · measured via OpenRouter routing · ✕ = probed, endpoint returned no measurable stream · SMOKE: of 5 programmatic capability probes passed (JSON schema, instruction following, tool call, long-context, benign compliance; retry-once) · screened via OpenRouter routing · v0.2 · methodology v1.1 · measured 24 hours ago
SERVE THIS MODEL AND NOT LISTED? LIST YOUR INFERENCE →
Quality benchmarksCURATED AUG 24, 2026
Third-party reported results, hand-curated with sources — not measured by ModelIndex. Our own measured serving data (TTFT/TPS/SMOKE) is in the providers table above.
Input price history · best listed price, observed
↓ 68% SINCE SEP 17, 2026
$2.10 · SEP 17, 2026$0.68 · TODAY
Pricing variants · USD / MTok in / out
STANDARD
$0.68 / $15.00
per MTok in / out
BATCH
$2.28 / $11.40
async batch
CACHE READ $0.45
Recent changesRSS ↗
Oct 8, 2026price increased · input0.615 → 0.68
Oct 7, 2026price increased · input0.365 → 0.615
Oct 6, 2026price decreased · input0.652 → 0.365
Oct 6, 2026price increased · output9 → 9.75
Oct 5, 2026price increased · input0.576 → 0.652
Oct 5, 2026price decreased · output11.25 → 9
Oct 4, 2026endpoint added · amazon-bedrock25 → 26
Oct 2, 2026price increased · input0.58 → 1.39
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 output943,718
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 API23 minutes ago
AGGREGATORHugging Face23 minutes ago
COMMUNITYModelIndex benchmarks19 days ago
MEASUREDVerified 23 minutes ago · list prices, not negotiated rates.