MiniMax M1 80K vs Qwen3 235B A22B
Wins 3 of 6 areas
Agents · Long documents · Following instructions
Wins 2 of 6 areas
Coding · Reasoning
MiniMax M1 80K wins more areas, narrowly.Qwen3 235B A22B is better at coding.
Scores updated · 30 tests both models report · How we compare
Where each one wins
Tests won in each of the six 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.
- Long documentsFinding answers in very long texts20MiniMax M1 80K2 of 2 tests
- Following instructionsDoing exactly what it is asked20MiniMax M1 80K2 of 2 tests
- AgentsCarrying out multi-step tasks on its own10MiniMax M1 80K1 of 1 test
- CodingWriting and fixing software12Qwen3 235B A22B2 of 3 tests
- ReasoningHard problems that need careful thinking01Qwen3 235B A22B1 of 3 tests · 2 ties
- FactsGetting facts right instead of making them up11Even1 each
Agents rests on a single test.
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.
MiniMax M1 80K costs 21% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where MiniMax M1 80K pulls ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+21.6points ahead
- Real work tasks from 44 professionsGDPVal+12.9points ahead
- Questions about very long textsLongBench v2+11.4points ahead
Where Qwen3 235B A22B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+13.9points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+3.1points ahead
- Code for real scientific research problemsSciCode+2.5points ahead
Every test, side by side
All 30 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3 235B A22B
- SWE-bench VerifiedMiniMax M1 80K by 21.65634.4+21.6
- Terminal-Bench HardQwen3 235B A22B by 3.136.1+3.1
- SciCodeQwen3 235B A22B by 2.537.439.9+2.5
ReasoningQwen3 235B A22B
- Humanity's Last ExamQwen3 235B A22B by 2.18.911+2.1
- GPQA Diamondtie69.770tie
- CritPttie00tie
FactsEven
- AA-Omniscience · Non-hallucinationQwen3 235B A22B by 13.98.522.4+13.9
- AA-Omniscience · AccuracyMiniMax M1 80K by 422.518.5+4
Long documentsMiniMax M1 80K
- AA-LCRMiniMax M1 80K by 57.757.70+57.7
- LongBench v2MiniMax M1 80K by 11.461.550.1+11.4
Following instructionsMiniMax M1 80K
- Multi-ChallengeMiniMax M1 80K by 3.544.741.2+3.5
- IFBenchMiniMax M1 80K by 3.141.838.7+3.1
Other results17 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.
- ZebraLogicMiniMax M1 80K by 49.186.837.7+49.1
- OpenAI-MRCR (128k)MiniMax M1 80K by 45.773.427.7+45.7
- TAU-bench (airline)MiniMax M1 80K by 27.36234.7+27.3
- MMLU-ProMiniMax M1 80K by 12.981.168.2+12.9
- AA Agentic IndexMiniMax M1 80K by 10.729.118.4+10.7
- τ²-Bench Telecom (AA run)MiniMax M1 80K by 10.234.224+10.2
- LiveCodeBenchQwen3 235B A22B by 5.76570.7+5.7
- MATH-500 (EM)MiniMax M1 80K by 5.696.891.2+5.6
- FullStackBenchMiniMax M1 80K by 5.468.362.9+5.4
- SimpleQAMiniMax M1 80K by 5.318.513.2+5.3
- TAU-bench (retail)MiniMax M1 80K by 4.963.558.6+4.9
- AIME 2025Qwen3 235B A22B by 4.676.981.5+4.6
- AA-OmniscienceQwen3 235B A22B by 3.8-48.5-44.7+3.8
- Artificial Analysis Coding IndexQwen3 235B A22B by 2.914.517.4+2.9
- AA IntelligenceMiniMax M1 80K by 2.211.79.5+2.2
- LiveCodeBench (24/8~25/5)tie6565.9tie
- AIME 2024tie8685.7tie
Questions people ask
Which is better, MiniMax M1 80K or Qwen3 235B A22B?
MiniMax M1 80K wins three of the six areas where both have results: agents, long documents and following instructions. Qwen3 235B A22B wins coding and reasoning. They are level on facts.
Which is better for coding?
Qwen3 235B A22B. It wins 2 of the 3 coding tests both models report; MiniMax M1 80K wins 1.
Which is cheaper?
MiniMax M1 80K costs $0.55 per million input tokens and $2.20 per million output tokens; Qwen3 235B A22B costs $0.70 and $2.80. That makes MiniMax M1 80K about 21% cheaper for the same work.
How do you compare the two?
We use the 30 benchmark tests both models have published scores on. The verdict counts the 13 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 17 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.