Moonshot AI·released 2025-07-121 source
Kimi K2 (Jul 2025) benchmark scores: 7 benchmarks tracked. 0% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 7 benchmarks·best result 85.6 on Lech Mazur Writing (unverified)·0 of 7 independently reproduced
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| Lech Mazur Writing | 85.6accuracy (%) | unverifiedT1 | 2025-07-12lechmazur/writing Github repository |
| Fiction.LiveBench | 61.1accuracy (%) | unverifiedT1 | 2025-07-12Fiction.live leaderboard |
| Aider polyglot | 59.1accuracy (%) | unverified· optimizedT1 | 2025-07-12Aider LLM Leaderboards |
| WeirdML | 39.36accuracy (%) | unverifiedT1 | 2025-07-12WeirdML Leaderboard |
| Terminal Bench | 27.8accuracy (%) | unverified· optimizedT1 | 2025-07-12Terminal-Bench v2 Leaderboard |
| SimpleBench | 11.56accuracy (%) | unverifiedT1 | 2025-07-12SimpleBench Leaderboard |
| GSO-Bench | 4.9accuracy (%) | unverifiedT1 | 2025-07-12https://gso-bench.github.io/index.html |
Claims drawn from cited facts, not live model generation.
This model was developed by Moonshot AI and released in 2025; no parameter count is offered, so its scale cannot be characterized here.
2 cited facts
This model is tracked across seven benchmarks. It holds no current top score on any of them. Because the record is unverified, no outside evaluation has confirmed or disputed these scores, so treat them as unconfirmed rather than verified.
3 cited facts
Its mean gap to the best recorded score across its tracked benchmarks is 46.66 points.
1 cited fact
1 facts cross-checked across data sources: 1 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: epoch_benchmarks
Kimi K2 (Jul 2025) is an AI model developed by Moonshot AI, released 2025-07-12. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
Kimi K2 (Jul 2025) has recorded scores on 7 benchmarks, each shown with its evidence status.
Kimi K2 (Jul 2025) has recorded scores on 7 benchmarks — Lech Mazur Writing, WeirdML, SimpleBench, Aider polyglot, GSO-Bench, Fiction.LiveBench, and 1 more. The full table above shows each score with its evidence status.
0 of 7 recorded scores (0%) are independently reproduced rather than self-reported by the lab.
Kimi K2 (Jul 2025) has 7 tracked claims: 7 unverified.