Moonshot AI model product
Kimi K3
Kimi K3 is Moonshot AI's flagship open Mixture-of-Experts model for long-horizon coding, knowledge work, and reasoning.
Updated Aug 17, 2026. Default version: Kimi K3
LLMBoard Score92.0Kimi K3
Coverage60%26 benchmark families
Context window1MTokens
Official input price$3Moonshot AI API
Kimi K3 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Kimi K3 Benchmark Results
Benchmark scores for Kimi K3.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkBabyVision | Score85.7% | Rank01 | Participants10 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkBrowseComp | Score91.2% | Rank01 | Participants62 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkDECK-Bench | Score73.5% | Rank01 | Participants1 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkDeepSearchQA | Score95.0% | Rank01 | Participants9 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkKimi Code Bench v2 | Score72.9% | Rank01 | Participants2 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkMathVision | Score97.8% | Rank01 | Participants33 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkMLS-Bench Lite | Score48.3% | Rank01 | Participants3 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkOmniDocBench | Score91.1% | Rank01 | Participants2 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkProgram Bench | Score77.8% | Rank01 | Participants6 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkSpreadsheetBench 2 | Score34.8% | Rank01 | Participants1 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkZEROBench | Score0.41 points | Rank01 | Participants9 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkAA-Briefcase | Score1,548 points | Rank02 | Participants3 | Percentile50.0% | EvidenceB | Evaluated |
| BenchmarkAPEX-Agents | Score37.6% | Rank02 | Participants8 | Percentile85.7% | EvidenceC | Evaluated |
| BenchmarkCharXiv-R | Score91.3% | Rank02 | Participants51 | Percentile98.0% | EvidenceC | Evaluated |
| BenchmarkFrontierSWE | Score81.2% | Rank02 | Participants16 | Percentile93.3% | EvidenceC | Evaluated |
| BenchmarkMCP Atlas | Score84.2% | Rank02 | Participants33 | Percentile96.9% | EvidenceC | Evaluated |
| BenchmarkMMMU-Pro (with tools) | Score83.4% | Rank02 | Participants4 | Percentile66.7% | EvidenceC | Evaluated |
| BenchmarkPerceptionBench | Score58.5% | Rank02 | Participants2 | Percentile0.0% | EvidenceC | Evaluated |
| BenchmarkSWE-Marathon | Score42.0% | Rank02 | Participants4 | Percentile66.7% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 2.1 | Score88.3% | Rank02 | Participants28 | Percentile96.3% | EvidenceC | Evaluated |
| BenchmarkAutomationBench | Score30.8% | Rank03 | Participants12 | Percentile81.8% | EvidenceC | Evaluated |
| BenchmarkDeepSWE | Score67.5% | Rank03 | Participants11 | Percentile80.0% | EvidenceC | Evaluated |
| BenchmarkJob Bench | Score52.9% | Rank03 | Participants4 | Percentile33.3% | EvidenceC | Evaluated |
| BenchmarkPostTrainBench | Score36.6% | Rank03 | Participants6 | Percentile60.0% | EvidenceC | Evaluated |
| BenchmarkToolathlon | Score73.2% | Rank03 | Participants37 | Percentile94.4% | EvidenceC | Evaluated |
| BenchmarkWorldVQA | Score51.0% | Rank03 | Participants5 | Percentile50.0% | EvidenceC | Evaluated |
| BenchmarkDeepSWE 1.1 | Score69.0% | Rank04 | Participants25 | Percentile87.5% | EvidenceB | Evaluated |
| BenchmarkOfficeQA Pro | Score63.3% | Rank04 | Participants7 | Percentile50.0% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score93.5% | Rank07 | Participants239 | Percentile97.5% | EvidenceC | Evaluated |
| BenchmarkMMMU-Pro | Score81.6% | Rank07 | Participants68 | Percentile91.0% | EvidenceC | Evaluated |
| BenchmarkHumanity's Last Exam | Score56.0% | Rank11 | Participants99 | Percentile89.8% | EvidenceC | Evaluated |
Kimi K3 Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenaagent task outcome explicit | Categoryoverall | Rank02 | Rating / score0.2 | VotesN/A | Observations23.2K | Result date |
| Arenaagent tool hallucination | Categoryoverall | Rank02 | Rating / score0.0 | VotesN/A | Observations2.3M | Result date |
| Arenawebdev | Categoryoverall | Rank02 | Rating / score1673.9 | Votes4,546 | ObservationsN/A | Result date |
| Arenaagent praise complaint | Categoryoverall | Rank04 | Rating / score0.2 | VotesN/A | Observations8.9K | Result date |
| Arenaagent | Categoryoverall | Rank05 | Rating / score0.1 | VotesN/A | Observations2.4M | Result date |
| Arenaagent steerability | Categoryoverall | Rank10 | Rating / score0.1 | VotesN/A | Observations28.5K | Result date |
| Arenatext style control | Categoryoverall | Rank11 | Rating / score1488.6 | Votes11,864 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank15 | Rating / score1476.3 | Votes11,864 | ObservationsN/A | Result date |
| Arenaagent bash recovery steps | Categoryoverall | Rank20 | Rating / score0.1 | VotesN/A | Observations66.2K | Result date |
| Arenatext factuality | Categoryoverall | Rank21 | Rating / score1471.0 | Votes11,864 | ObservationsN/A | Result date |
Kimi K3 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $3 input, $15 output per 1M
- Official provider
- Moonshot AI
- Lowest third-party
- From $2 input, $8 output per 1M via CrofAI
- Tracked offerings
- 54
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderUmans AI Coding Plan | Provider model IDumans-kimi-k3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderKenari | Provider model IDkimi-k3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderSCNet Token Plan | Provider model IDKimi-K3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderKimi For Coding | Provider model IDk3 | Regionglobal | Input / 1MN/A | Output / 1MN/A | Context1M | Updated |
| ProviderCrofAI | Provider model IDkimi-k3 | Regionglobal | Input / 1M$2 | Output / 1M$8 | Context1M | Updated |
| ProviderRequesty | Provider model IDkimi-k3 | Regionglobal | Input / 1M$2.25 | Output / 1M$11.25 | Context1M | Updated |
| ProviderNanoGPT | Provider model IDmoonshotai/kimi-k3 | Regionglobal | Input / 1M$2.5 | Output / 1M$13.5 | Context1M | Updated |
| ProviderDeep Infra | Provider model IDmoonshotai/Kimi-K3 | Regionglobal | Input / 1M$2.85 | Output / 1M$14.25 | Context1M | Updated |
| ProviderDigitalOcean | Provider model IDkimi-k3 | Regionglobal | Input / 1M$2.85 | Output / 1M$14.25 | Context1M | Updated |
| ProviderNeuralwatt | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderCortecs | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderSynthetic | Provider model IDhf:moonshotai/Kimi-K3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context524.3K | Updated |
| ProviderHugging Face | Provider model IDmoonshotai/Kimi-K3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderImpossibl | Provider model IDmoonshotai/kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderOpenRouter | Provider model IDmoonshotai/kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderMoonshot AI | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderCrossModel | Provider model IDmoonshot/kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderOpenCode Go | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderVivgrid | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderNebius Token Factory | Provider model IDmoonshotai/Kimi-K3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderMerge Gateway | Provider model IDmoonshot/kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderBerget.AI | Provider model IDmoonshotai/Kimi-K3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context327.7K | Updated |
| ProviderZenMux | Provider model IDmoonshotai/kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderPerplexity Agent | Provider model IDmoonshot-ai/kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderUmans AI | Provider model IDumans-kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderTogether AI | Provider model IDmoonshotai/Kimi-K3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderOfox | Provider model IDmoonshotai/kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderNeon | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderOpenCode Zen | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderCharm Hyper | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderBaseten | Provider model IDmoonshotai/Kimi-K3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderModal | Provider model IDmoonshotai/Kimi-K3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderAIHubMix | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderClinePass | Provider model IDcline-pass/kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderEmpirioLabs AI | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderVercel AI Gateway | Provider model IDmoonshotai/kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderEden AI | Provider model IDmoonshot/kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderNovitaAI | Provider model IDmoonshotai/kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderLLM Gateway | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderCoralBricks | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| Providerai& | Provider model IDmoonshotai/kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$12.5 | Context1M | Updated |
| ProviderGitHub Copilot | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderTensorX | Provider model IDmoonshotai/kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderKilo Gateway | Provider model IDmoonshotai/kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderPioneer | Provider model IDmoonshotai/Kimi-K3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderFireworks AI | Provider model IDaccounts/fireworks/models/kimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderMoonshot AI (China) | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderWeights & Biases | Provider model IDmoonshotai/Kimi-K3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderAbacus | Provider model IDmoonshotai/Kimi-K3 | Regionglobal | Input / 1M$3 | Output / 1M$15 | Context1M | Updated |
| ProviderVenice AI | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3.75 | Output / 1M$18.75 | Context1M | Updated |
| ProviderGreenPT | Provider model IDkimi-k3 | Regionglobal | Input / 1M$3.76 | Output / 1M$18.81 | Context1M | Updated |
| ProviderTinfoil | Provider model IDkimi-k3 | Regionglobal | Input / 1M$4 | Output / 1M$20 | Context262.1K | Updated |
| ProviderLLM Gateway | Provider model IDkimi-k3-fast | Regionglobal | Input / 1M$4.5 | Output / 1M$22.5 | Context1M | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Kimi K3 Runtime Performance
Provider-specific output speed and catalog latency for Kimi K3. Runtime does not affect the capability score.
Sort by
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderFireworks | Output Speed25.734 tok/s | Catalog Latency4.248 s | Max Input1M | Max Output1M | Updated |
| ProviderTogether | Output Speed23.56 tok/s | Catalog Latency13.029 s | Max Input1M | Max Output1M | Updated |
| ProviderMoonshot AI | Output Speed4.128 tok/s | Catalog Latency6.861 s | Max Input1M | Max Output1M | Updated |
| ProviderNovita | Output Speed0.196 tok/s | Catalog Latency9.84 s | Max Input1M | Max Output1M | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Kimi K3 Specifications
Technical details for the model's default version.
- Version
- Kimi K3
- Released
- Jul 16, 2026
- Knowledge cutoff
- Unknown
- Parameters
- 2.8T
- Context window
- 1M
- Max output
- 131.1K
- Inputs
- image, text, video
- Outputs
- text
- Open weights
- Yes
- License
- Kimi K3 License
Kimi K3 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionKimi K3 | Released | LLMBoard92.0 | Parameters2.8T | Context1M | Max output131.1K | Open weightsYes | LicenseKimi K3 License |
Kimi K3 vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to Kimi K3
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
MA
Kimi K2.6
Moonshot AI
MA
Kimi K2.7 Code
Moonshot AI
MA
Kimi K2.5
Moonshot AI
MA
Kimi K2 Thinking
Moonshot AI
MA
Kimi K2
Moonshot AI
MA
Kimi k1.5
Moonshot AI
What is Kimi K3?
Key information about Kimi K3 and its available data.
Kimi K3 is Moonshot AI's flagship open Mixture-of-Experts model for long-horizon coding, knowledge work, and reasoning. 8 trillion total parameters, activates 16 of 896 experts, and combines Kimi Delta Attention, Attention Residuals, and Stable LatentMoE.
The model supports a 1M-token context window, native image and video understanding, tool calling, structured output, and always-on reasoning with max thinking effort at launch.
Data as of 2026-08-17.
FAQ
Common questions about Kimi K3.
When was Kimi K3 released?
Kimi K3's default version was released on Jul 16, 2026.
How much does Kimi K3 cost?
Kimi K3's official API price is $3 per million input tokens and $15 per million output tokens via Moonshot AI. The lowest tracked third-party offer starts at $2 input and $8 output via CrofAI.
Who created Kimi K3?
Kimi K3 was created by Moonshot AI.
What is the context window for Kimi K3?
The default version has a 1M token context window.
Is Kimi K3 open weight?
Yes. The default version is marked as open weight under Kimi K3 License.
How many API providers offer Kimi K3?
54 provider offerings are linked to the default version.
What models should I compare Kimi K3 with?
Nearby ranked alternatives include GPT-5.6-Sol, Claude Fable 5, Claude Mythos.