Kimi K3: benchmarks, price and rank
Kimi K3 from Moonshot is 25th of 123 rated AI models on Spezia's leaderboard, strongest on BrowseComp and MCP Atlas. It costs $2.85 in, $14 out per million tokens.
- 25th of 123 rated models
- Rating 1,537
- First on 1 of 17 benchmarks
- Open weights
- Released Jul 16, 2026
- 1 million tokens of context
Where it is strongest
- 3rd of 56 on BrowseComp
- 3rd of 33 on MCP Atlas
- 1st of 45 on CharXiv Reasoning
- 1st of 8 on DeepSearchQA
Where it falls behind
- 14th of 21 on FrontierSWE
- 35th of 56 on APEX-Agents
- 8th of 13 on OfficeQA Pro
Price: $2.85 in, $14 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.
- APEX-Agents47.8%
Long tasks written by investment banking analysts, management consultants and corporate lawyers from their own work, done across files and apps in a realistic work setup.
35th of 56. Independent result from Epoch AI.
- BrowseComp91.2%
Questions whose answers are hidden deep on the web and take many searches and pages to track down.
3rd of 56. The maker's own number, from Moonshot AI, via LLM Stats.
- MCP Atlas84.2%
Multi-step tasks where the model has to pick and combine many real tools, connected through MCP, the standard way agents plug into apps and data.
3rd of 33. The maker's own number, from Moonshot AI, via LLM Stats.
- FrontierSWE25.9%
Open-ended technical projects that take hours to tens of hours, from making systems faster to building large programs and applied machine learning research. Each run gets a 20-hour budget.
14th of 21. Independent result from Epoch AI.
- Humanity's Last Exam56%
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.
10th of 74. The maker's own number, from Moonshot AI, via LLM Stats.
- Vending-Bench 2$5,165
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.
32nd of 60. Independent result from Epoch AI.
- Toolathlon73.2%
Long tasks across many real apps and their tools, such as email, calendars, spreadsheets, code repositories and online stores.
6th of 37. The maker's own number, from Moonshot AI, via LLM Stats.
- DeepSWE 1.169%
Software engineering tasks run in the same simple agent setup for every model.
Independent result from LLM Stats.
- SimpleBench60.7%
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.
39th of 68. Independent result from Epoch AI.
- SimpleQA Verified50.6%
Short factual questions with one checked answer, asked without web search.
30th of 78. Independent result from Epoch AI.
12 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.