MiMo V2.5 vs MiMo V2.5 Pro
Wins 0 of 6 areas
—
Wins 6 of 6 areas
Coding · Agents · Reasoning · Facts · Long documents · Following instructions
MiMo V2.5 Pro is the stronger all-rounder.MiMo V2.5 is cheaper.
Scores updated · 37 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.
- CodingWriting and fixing software07MiMo V2.5 Pro7 of 7 tests
- AgentsCarrying out multi-step tasks on its own02MiMo V2.5 Pro2 of 3 tests · 1 tie
- ReasoningHard problems that need careful thinking02MiMo V2.5 Pro2 of 3 tests · 1 tie
- FactsGetting facts right instead of making them up02MiMo V2.5 Pro2 of 2 tests
- Long documentsFinding answers in very long texts01MiMo V2.5 Pro1 of 1 test
- Following instructionsDoing exactly what it is asked01MiMo V2.5 Pro1 of 1 test
Long documents and following instructions 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.
MiMo V2.5 costs 68% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where MiMo V2.5 Pro pulls ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+48.1points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+12.8points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+8.5points ahead
Where MiMo V2.5 pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 37 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingMiMo V2.5 Pro
- SWE-bench VerifiedMiMo V2.5 Pro by 48.130.878.9+48.1
- SciCodeMiMo V2.5 Pro by 6.743.950.6+6.7
- LiveCodeBench v6MiMo V2.5 Pro by 4.135.539.6+4.1
- LMArena · WebDevMiMo V2.5 Pro by 41 rating points14381479+41 rating
- Terminal-Bench 2.1MiMo V2.5 Pro by 1.563.765.2+1.5
- Terminal-Bench HardMiMo V2.5 Pro by 1.541.743.2+1.5
- SWE-bench ProMiMo V2.5 Pro by 1.156.157.2+1.1
AgentsMiMo V2.5 Pro
- GDPValMiMo V2.5 Pro by 6.124.931+6.1
- τ-Bench V3 · BankingMiMo V2.5 Pro by 1.28.79.9+1.2
- Terminal-Bench 4.0tie00tie
ReasoningMiMo V2.5 Pro
- Humanity's Last ExamMiMo V2.5 Pro by 8.527.235.7+8.5
- GPQA DiamondMiMo V2.5 Pro by 1.784.986.6+1.7
- CritPttie3.74tie
FactsMiMo V2.5 Pro
- AA-Omniscience · Non-hallucinationMiMo V2.5 Pro by 7.268.175.3+7.2
- AA-Omniscience · AccuracyMiMo V2.5 Pro by 5.616.822.4+5.6
Following instructionsMiMo V2.5 Pro
- IFBenchMiMo V2.5 Pro by 12.867.179.9+12.8
Other results20 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.
- MATHMiMo V2.5 Pro by 18.567.786.2+18.5
- AA-OmniscienceMiMo V2.5 Pro by 13.1-9.83.3+13.1
- GlobalMMLUMiMo V2.5 Pro by 6.277.483.6+6.2
- Finance Agent v2MiMo V2.5 Pro by 4.836.741.5+4.8
- AA IntelligenceMiMo V2.5 Pro by 4.221.826+4.2
- τ²-Bench Telecom (AA run)MiMo V2.5 Pro by 3.690.694.2+3.6
- MBPP+MiMo V2.5 Pro by 3.270.974.1+3.2
- MMLUMiMo V2.5 Pro by 3.186.389.4+3.1
- MMLU-ReduxMiMo V2.5 Pro by 389.892.8+3
- MMLU-ProMiMo V2.5 Pro by 2.765.868.5+2.7
- DROPMiMo V2.5 Pro by 2.683.786.3+2.6
- Terminal-Bench 2.0MiMo V2.5 Pro by 2.665.868.4+2.6
- CMMLUMiMo V2.5 Pro by 288.290.2+2
- HellaSwagMiMo V2.5 Pro by 1.288.689.8+1.2
- WinoGrandetie84.785.6tie
- Claw Eval (pass@3)tie63.264tie
- ARC-Challengetie96.597.2tie
- TriviaQAtie80.781.3tie
- aime_2024_2025_combinedtie36.937.3tie
- Toolathlon Verifiedtie49.149.1tie
Questions people ask
Which is better, MiMo V2.5 or MiMo V2.5 Pro?
MiMo V2.5 Pro wins all six areas where both have results: coding, agents, reasoning, facts, long documents and following instructions. MiMo V2.5 wins none, but costs 68% less.
Which is better for coding?
MiMo V2.5 Pro. It wins 7 of the 7 coding tests both models report; MiMo V2.5 wins none.
Which is cheaper?
MiMo V2.5 costs $0.11 per million input tokens and $0.22 per million output tokens; MiMo V2.5 Pro costs $0.35 and $0.70. That makes MiMo V2.5 about 68% cheaper for the same work.
How do you compare the two?
We use the 37 benchmark tests both models have published scores on. The verdict counts the 17 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 20 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.