DeepSeek-V4-Pro-Max: benchmarks, price and rank
DeepSeek-V4-Pro-Max from DeepSeek is 57th of 123 rated AI models on Spezia's leaderboard, strongest on SWE-bench Verified and MMLU-Pro. It costs $1.30 in, $2.60 out per million tokens.
- 57th of 123 rated models
- Rating 1,275
- Rated on 7 benchmarks
- Open weights
- Released Apr 23, 2026
- 1 million tokens of context
Where it is strongest
- 4th of 56 on SWE-bench Verified
- 6th of 56 on MMLU-Pro
Where it falls behind
- 33rd of 50 on SWE-Bench Pro
Price: $1.30 in, $2.60 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.
- BrowseComp83.4%
Questions whose answers are hidden deep on the web and take many searches and pages to track down.
18th of 56. The maker's own number, from DeepSeek, via LLM Stats.
- MCP Atlas73.6%
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.
16th of 33. The maker's own number, from DeepSeek, via LLM Stats.
- SWE-Bench Pro55.4%
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.
33rd of 50. The maker's own number, from DeepSeek, via LLM Stats.
- Humanity's Last Exam48.2%
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.
20th of 74. The maker's own number, from DeepSeek, via LLM Stats.
- Toolathlon51.8%
Long tasks across many real apps and their tools, such as email, calendars, spreadsheets, code repositories and online stores.
17th of 37. The maker's own number, from DeepSeek, via LLM Stats.
- IMO-AnswerBench89.8%
Problems from the International Mathematical Olympiad, the hardest contest for school students, where the model has to reach the right final answer.
4th of 18. The maker's own number, from DeepSeek, via LLM Stats.
- SWE-bench Verified80.6%
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.
4th of 56. The maker's own number, from DeepSeek, via LLM Stats.
- GPQA Diamond90.1%
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 DeepSeek, via LLM Stats.
- MMLU-Pro87.5%
Twelve thousand multiple-choice questions across fourteen subjects, from law and medicine to engineering, with ten options each.
6th of 56. The maker's own number, from DeepSeek, via LLM Stats.
Its full profile is in Spezia.
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.