MiniMax M2.7 vs Solar Pro4
Wins 2 of 6 areas
Coding · Long documents
Wins 3 of 6 areas
Agents · Reasoning · Math
Solar Pro4 wins more areas, narrowly.MiniMax M2.7 is better at coding.
Scores updated · 21 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.
- ReasoningHard problems that need careful thinking02Solar Pro42 of 3 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own12Solar Pro42 of 4 tests · 1 tie
- MathCompetition and research-level math01Solar Pro41 of 1 test
- CodingWriting and fixing software31MiniMax M2.73 of 4 tests
- Long documentsFinding answers in very long texts10MiniMax M2.71 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
Math and long documents rest on a single test each.
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 Solar Pro4 pulls ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+13.4points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+11.2points ahead
- US invitational high-school math exam problemsAIME 2026+5.5points ahead
Where MiniMax M2.7 pulls ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+9.3points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+7.9points ahead
- Code for real scientific research problemsSciCode+5.5points ahead
Every test, side by side
All 21 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingMiniMax M2.7
- SWE-bench VerifiedMiniMax M2.7 by 9.379.970.6+9.3
- SciCodeMiniMax M2.7 by 5.550.144.6+5.5
- LMArena · WebDevMiniMax M2.7 by 27 rating points13981371+27 rating
- Terminal-Bench 2.1Solar Pro4 by 1.955.457.3+1.9
AgentsSolar Pro4
- τ-Bench V3 · BankingSolar Pro4 by 13.49.923.3+13.4
- BrowseCompMiniMax M2.7 by 4.954.149.2+4.9
- GDPValSolar Pro4 by 4.825.730.5+4.8
- Terminal-Bench 4.0tie00.5tie
ReasoningSolar Pro4
- CritPtSolar Pro4 by 4.80.65.4+4.8
- GPQA DiamondSolar Pro4 by 1.787.489.1+1.7
- Humanity's Last Examtie29.629.2tie
FactsEven
- AA-Omniscience · Non-hallucinationSolar Pro4 by 11.264.475.6+11.2
- AA-Omniscience · AccuracyMiniMax M2.7 by 7.926.818.9+7.9
Other results6 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.
- AA Agentic IndexSolar Pro4 by 16.816.833.6+16.8
- τ³-Bench BankingSolar Pro4 by 14.18.923+14.1
- AA IntelligenceSolar Pro4 by 5.422.828.2+5.4
- MMLU-ProSolar Pro4 by 4.481.986.3+4.4
- AA-OmniscienceMiniMax M2.7 by 1.60.8-0.8+1.6
- Artificial Analysis Coding Indextie52.652.7tie
Questions people ask
Which is better, MiniMax M2.7 or Solar Pro4?
Solar Pro4 wins three of the six areas where both have results: agents, reasoning and math. MiniMax M2.7 wins coding and long documents. They are level on facts.
Which is better for coding?
MiniMax M2.7. It wins 3 of the 4 coding tests both models report; Solar Pro4 wins 1.
Which is cheaper?
MiniMax M2.7 costs $0.30 per million input tokens and $1.20 per million output tokens; Solar Pro4 costs $0.30 and $1.20. For a million tokens read plus a million written, they cost about the same.
How do you compare the two?
We use the 21 benchmark tests both models have published scores on. The verdict counts the 15 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 6 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.