MiMo-V2.5-Pro
Xiaomi·Released Apr 22, 2026Open weights
Updated Oct 8, 9:18 PM ET
The score is on the board's scale (reference average 50, about 15 points per standard deviation), not a percentage.
By domain
Each domain is scored on the same scale as the overall score. Coverage is the share of the domain's kinds of evaluation with a result; it says how much evidence there is, not how good the model is.
Every result behind the score
The result as the evaluator publishes it, and the reading it gives on the board's scale. A hard evaluation can give a high reading for a modest result; results near 95% or 5% only say "at least" or "at most".
Coding
5 results · 41.7- SciCodeArtificial Analysis · thinking effort50.6%Reading 51.2
- Code MigrationVals AI21.6%Reading 43.4
- Vibe Code BenchVals AI34.1%Reading 38.8
- Terminal-Bench 4.0 (AA run)Artificial Analysis · thinking effortNear the floor0.0%Reading at most 34.5
- Terminal-Bench 4.0Vals AINear the floor0.5%Reading at most 33.0
Research and reasoning
2 results · 45.2- Humanity's Last Exam (AA run)Artificial Analysis · thinking effort35.7%Reading 50.7
- ProofBenchVals AI22.0%Reading 41.9
Professional work
8 results · 30.3- τ²-Bench Telecom (AA run)Artificial Analysis · thinking effort94.2%Reading 56.8
- EMBVals AI55.2%Reading 51.5
- Finance AgentVals AI41.5%Reading 37.0
- Legal Research BenchVals AI15.9%Reading 27.4
- Tax Agent BenchVals AI11.5%Reading 27.3
- Harvey Legal Agent BenchmarkVals AINear the floor2.1%Reading at most 24.8
- τ-Bench Banking (AA run)Artificial Analysis · thinking effort9.9%Reading 7.1
- MedCodeVals AI32.5%Reading 6.7
Knowledge and accuracy
3 results · 58.4- BullshitBenchBullshitBench · xhigh effort45.5%Reading 70.3
- IFBench (AA run)Artificial Analysis · thinking effort79.9%Reading 65.1
- AA-LCRArtificial Analysis · thinking effort79.7%Reading 48.5
Human preference
1 result · 44.1- Arena TextLMArena1,467.5Reading 45.7
Shown, not scored
Composite indices, saturated or older evaluations, evaluations still being checked, and vision, writing and multilingual results. They do not move the score.
- LiveCodeBenchVals AI · CodingReference only81.4%
- SWE-bench VerifiedVals AI · CodingReference only74.0%
- CorpFinVals AI · Professional workReference only61.4%
- LegalBenchVals AI · Professional workReference only77.1%
- MedScribeVals AI · Professional workWatching83.7%
- TaxEvalVals AI · Professional workReference only73.8%
- GPQA DiamondVals AI · Knowledge and accuracyReference only82.6%
- MMLU-ProVals AI · Knowledge and accuracyReference only84.6%
- AA Intelligence IndexArtificial Analysis · Composite indicesReference only26.0
- Vals IndexVals AI · Composite indicesReference only33.8%
- Arena WebDevLMArena · Writing and designReference only1,478.4
No result yet
Scored evaluations this model has not taken. A missing result neither adds nor subtracts.
- LiveBench Coding · LiveBench
- ProgramBench · Vals AI
- Vibe Code Bench 1–100 · Vals AI
- CyberBench Patch · Vals AI
- APEX-SWE · Mercor
- FrontierMath Tiers 1–3 · Epoch AI
- FrontierMath Tier 4 · Epoch AI
- LiveBench Reasoning · LiveBench
- Chess Puzzles · Epoch AI
- Mystery Game Puzzles · Epoch AI
- MysteryMechanism · Vals AI
- Terminal-Bench Science · Vals AI
- APEX-Agents · Mercor
- LiveBench Data Analysis · LiveBench
- SimpleQA Verified · Epoch AI
- LiveBench Language · LiveBench
- LiveBench Instruction Following · LiveBench