Kimi K2-Thinking-0905: benchmarks, price and rank
Kimi K2-Thinking-0905 from Moonshot is 116th of 123 rated AI models on Spezia's leaderboard. It costs $0.47 in, $2 out per million tokens.
- 116th of 123 rated models
- Rating 460
- Rated on 7 benchmarks
- Open weights
- Released Jul 11, 2025
- 131,000 tokens of context
Where it is strongest
Not in the top tenth on any benchmark in use.
Where it falls behind
- 74th of 74 on Humanity's Last Exam
- 17th of 18 on IMO-AnswerBench
- 64th of 68 on SimpleBench
Price: $0.47 in, $2 out per million tokens.
Its headline results
The benchmarks that count most in the rating, with its place among the models measured the same way, and where each number came from.
- BrowseComp60.2%
Questions whose answers are hidden deep on the web and take many searches and pages to track down.
39th of 56. The maker's own number, from Moonshot AI, via LLM Stats.
- Humanity's Last Exam4.7%
About 2,500 questions written by experts in more than a hundred fields to be as hard as they could make them, many needing a diagram or image.
74th of 74. The maker's own number, from Moonshot AI, via LLM Stats.
- LiveCodeBench v653.7%
Programming contest problems collected after models were trained, from LeetCode, AtCoder and Codeforces.
39th of 43. The maker's own number, from Moonshot AI, via LLM Stats.
- SimpleBench39.6%
Multiple-choice puzzles about everyday situations, space, time and people, written to trip up models that pattern-match instead of think. Ordinary adults do well on them.
64th of 68. Independent result from Epoch AI.
- LMArena text1,380
People put a question to two anonymous models and vote for the better answer.
94th of 107. Independent result from LMArena.
- IMO-AnswerBench78.6%
Problems from the International Mathematical Olympiad, the hardest contest for school students, where the model has to reach the right final answer.
17th of 18. The maker's own number, from Moonshot AI, via LLM Stats.
- HealthBench58%
5,000 realistic health conversations with patients and clinicians, each graded against criteria written by 262 physicians.
7th of 16. The maker's own number, from Moonshot AI, via LLM Stats.
- SciCode44.8%
Research coding problems written by scientists across 16 fields, from physics and chemistry to biology and materials science.
11th of 24. The maker's own number, from Moonshot AI, via LLM Stats.
- SWE-bench Verified71.3%
Real bug reports and feature requests from popular open-source Python projects. The model has to change the code so the project's hidden tests pass.
30th of 56. The maker's own number, from Moonshot AI, via LLM Stats.
- GPQA Diamond84.5%
Graduate-level questions in biology, physics and chemistry, written by PhDs so that searching the web does not give the answer away.
The maker's own number, from Moonshot AI, via LLM Stats.
5 more results in its full profile.
Every result with its source, what it costs per task, what jumped with its release, side by side with any other model, and answers to your questions about when to use it.
Models ranked near it
How the rating works, and every benchmark it uses: the model leaderboard.