Kimi K2.5: benchmarks, price and rank
Kimi K2.5 from Moonshot is 76th of 123 rated AI models on Spezia's leaderboard, strongest on AA-LCR. It costs $0.60 in, $3 out per million tokens.
- 76th of 123 rated models
- Rating 1,020
- Rated on 13 benchmarks
- Open weights
- Released Jan 27, 2026
- 262,000 tokens of context
Where it is strongest
- 3rd of 19 on AA-LCR
Where it falls behind
- 15th of 15 on CyberGym
- 48th of 50 on SWE-Bench Pro
- 51st of 60 on Vending-Bench 2
Price: $0.60 in, $3 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.
- BrowseComp74.9%
Questions whose answers are hidden deep on the web and take many searches and pages to track down.
28th of 56. The maker's own number, from Moonshot AI, via LLM Stats.
- SWE-Bench Pro50.7%
Long changes to real codebases in several languages, often across many files, including code from private and copyleft projects that models are unlikely to have seen.
48th of 50. The maker's own number, from Moonshot AI, via LLM Stats.
- Humanity's Last Exam24.4%
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.
Independent result from Epoch AI.
- Vending-Bench 2$1,198
The model runs a simulated vending machine business for a year on its own: finding suppliers, ordering stock, setting prices and paying the daily fees.
51st of 60. Independent result from Epoch AI.
- LiveCodeBench v685%
Programming contest problems collected after models were trained, from LeetCode, AtCoder and Codeforces.
12th of 43. The maker's own number, from Moonshot AI, via LLM Stats.
- SimpleBench46.8%
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.
57th of 68. Independent result from Epoch AI.
- SimpleQA Verified34.3%
Short factual questions with one checked answer, asked without web search.
55th of 78. Independent result from Epoch AI.
- MMMU-Pro78.5%
College exam questions that need a picture to answer, such as charts, diagrams, maps and chemical structures, with more answer options than the original MMMU and some questions shown only as a screenshot.
19th of 51. The maker's own number, from Moonshot AI, via LLM Stats.
- CharXiv Reasoning77.5%
Questions about real charts from scientific papers that need reasoning across the chart, not just reading off a value.
32nd of 45. The maker's own number, from Moonshot AI, via LLM Stats.
- AA-LCR70%
Questions that need reasoning across several long documents at once, about 100,000 tokens of text in all.
3rd of 19. The maker's own number, from Moonshot AI, via LLM Stats.
13 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.