Ctrl K
Back to models

Moonshot AI

Kimi-K3

  • Released
  • Open weights
  • Kimi K3 License
Compare
Intelligence Index

Max setting

43.6#15 of 83measured models
Measured · Artificial Analysis v4.3.2
API price from
$2.85 in / $14.25 out / 1M tokOpenRouter · DeepInfra · BF16
Context window
1.05M tok

Performance profile

ReasoningRank #6 of 107
93.5%
GPQA Diamond
CodingRank #9 of 94
59.5%
SciCode
Cost efficiencyRank #57 of 67
21.8pts / $
Intelligence ÷ eval cost

Capabilities

Context window

Modalities

Reasoning modes

Avg reasoning tokens / task — Artificial Analysis Intelligence Index v4.3.2: weighted average reasoning tokens per benchmark task, not a token limit. Source (opens in a new tab)

Integration

Reasoning trade-offs

Source (opens in a new tab)
Intelligence Index v4.3.2
Kimi-K3
Cost per task (USD)
Kimi-K3Max
Intelligence Index
43.6
Cost / task
$2.00
Evaluation time
17.9m

Pricing & access

API route
OpenRouter · DeepInfra · BF16API
Tier & contextStandard

USD per 1M tokens

Input
$2.85
Output
$14.25
Cached input
$0.285

Run locally

Original download
1,560.9 GB
App memory
Unverified
Documented software
vLLMSetup guide (opens in a new tab)
Find GGUF downloads (opens in a new tab)Original weights (opens in a new tab)

Benchmarks

Full leaderboardSource (opens in a new tab)

Humanity’s Last Exam

Frontier reasoning

Kimi-K3 · Max#13 of 126

  1. 1Claude Opus 5.5Max61.4%
  2. 2Claude Fable 5.1Max59.1%
  3. 3Gemini 4 ArgonHigh57.1%
  5. 10Muse Spark 1.3Max48.7%
  6. 11Gemini 3.8 FlashHigh47.8%
  7. 12Gemini 3.1 Pro PreviewReasoning47.0%
  8. 13Kimi-K3Max46.9%
  9. 14Step 5 PreviewReasoning46.5%
  10. 15Claude Haiku 5.5Max44.4%
  11. 16Muse Spark OpenExtra high43.9%

Peers use their best measured result on this benchmark.

Alternatives

Each model at its highest-scoring measured setting · Intelligence Index v4.3.2 and its evaluation cost per task.