Kimi K3
Moonshot AI · 2026-07-16
Basic Information
- Developer
- Moonshot AI
- Release Date
- 2026-07-16
- Overall Capability Index (ECI)
- 157.6
Benchmark Scores (Epoch AI)
Each benchmark also shows the model's rank among the models evaluated.
PhD-level science questions (GPQA Diamond) · #18 of 30
93.1%
Advanced mathematics (FrontierMath) · #21 of 30
72.2%
Reasoning through unfamiliar puzzles (ARC-AGI-2) · #24 of 30
60.4%
Factual question accuracy (SimpleQA Verified) · #24 of 30
50.6%
Bars show accuracy (%) and start at 0%.
User Vote Rankings (LMArena)
Rankings based on people's votes comparing responses from two models. Models do not appear on leaderboards where they have too few votes.
- Overall#16 of 413 · 1488
- English#15 of 413 · 1495
- Japanese#8 of 274 · 1502
- Coding#9 of 408 · 1541
- Korean#9 of 278 · 1473
- Spanish#9 of 295 · 1485
- Agent task success#9 of 50 · 0.076
- Chinese#10 of 389 · 1540
- German#13 of 307 · 1492
- French#13 of 290 · 1509
- Web development#13 of 50 · 1658
- Overall agent performance#14 of 50 · 0.042
- Math#16 of 396 · 1500
- Creative writing#26 of 411 · 1460
Where to Use It
This site has not yet compiled the services and plans that offer this model. Check the developer's official information.
Scores and rankings are reproduced directly from public sources and were not measured by this site. Model names vary slightly across sources and are matched automatically, so different versions may occasionally be mixed together.
Source: Epoch AI (CC BY 4.0) · Checked 2026-10-04 · LMArena (CC BY 4.0) · Checked 2026-10-04