DeepSeek-R1 0528 vs MiniMax M1 40K
Wins 2 of 4 areas
Coding · Reasoning
Wins 2 of 4 areas
Long documents · Following instructions
The two are evenly matched.DeepSeek-R1 0528 is better at reasoning and coding; MiniMax M1 40K at long documents and following instructions.
Scores updated · 24 tests both models report · How we compare
Where each one wins
Tests won in each of the four areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- ReasoningHard problems that need careful thinking20DeepSeek-R1 05282 of 2 tests
- CodingWriting and fixing software21DeepSeek-R1 05282 of 3 tests
- Long documentsFinding answers in very long texts01MiniMax M1 40K1 of 2 tests · 1 tie
- Following instructionsDoing exactly what it is asked01MiniMax M1 40K1 of 2 tests · 1 tie
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where DeepSeek-R1 0528 pulls ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+13.6points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+13.1points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+8points ahead
Where MiniMax M1 40K pulls ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+11points ahead
- Questions about very long textsLongBench v2+8.9points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+1.6points ahead
Every test, side by side
All 24 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingDeepSeek-R1 0528
- Terminal-Bench HardDeepSeek-R1 0528 by 13.615.92.3+13.6
- SWE-bench VerifiedMiniMax M1 40K by 1144.655.6+11
- SciCodeDeepSeek-R1 0528 by 2.540.337.8+2.5
ReasoningDeepSeek-R1 0528
- GPQA DiamondDeepSeek-R1 0528 by 13.181.368.2+13.1
- Humanity's Last ExamDeepSeek-R1 0528 by 815.87.8+8
Long documentsMiniMax M1 40K
- LongBench v2MiniMax M1 40K by 8.952.161+8.9
- AA-LCRtie55.755.7tie
Following instructionsMiniMax M1 40K
- IFBenchMiniMax M1 40K by 1.639.641.2+1.6
- Multi-Challengetie4544.7tie
Other results15 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- SimpleQADeepSeek-R1 0528 by 74.492.317.9+74.4
- OpenAI-MRCR (128k)MiniMax M1 40K by 24.651.576.1+24.6
- ZebraLogicDeepSeek-R1 0528 by 1595.180.1+15
- AIME 2025DeepSeek-R1 0528 by 12.987.574.6+12.9
- LiveCodeBenchDeepSeek-R1 0528 by 1173.362.3+11
- LiveCodeBench (24/8~25/5)DeepSeek-R1 0528 by 10.873.162.3+10.8
- Artificial Analysis Coding IndexDeepSeek-R1 0528 by 9.92414.1+9.9
- AIME 2024DeepSeek-R1 0528 by 8.191.483.3+8.1
- TAU-bench (airline)MiniMax M1 40K by 6.553.560+6.5
- τ²-Bench Telecom (AA run)DeepSeek-R1 0528 by 4.936.531.6+4.9
- MMLU-ProDeepSeek-R1 0528 by 4.48580.6+4.4
- TAU-bench (retail)MiniMax M1 40K by 3.963.967.8+3.9
- AA IntelligenceDeepSeek-R1 0528 by 3.113.110+3.1
- MATH-500 (EM)DeepSeek-R1 0528 by 29896+2
- FullStackBenchDeepSeek-R1 0528 by 1.869.467.6+1.8
Questions people ask
Which is better, DeepSeek-R1 0528 or MiniMax M1 40K?
DeepSeek-R1 0528 and MiniMax M1 40K each win two of the four areas where both have results. DeepSeek-R1 0528 wins coding and reasoning; MiniMax M1 40K wins long documents and following instructions.
Which is better for coding?
DeepSeek-R1 0528. It wins 2 of the 3 coding tests both models report; MiniMax M1 40K wins 1.
How do you compare the two?
We use the 24 benchmark tests both models have published scores on. The verdict counts the 9 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 15 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.