MiMo-V2.6-Flash
Xiaomi·Released Sep 21, 2026Open weights
Updated Oct 8, 21:18 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
7 results · 61.0- CyberBench PatchVals AI85.7%Reading 68.5
- ProgramBenchVals AINear the floor0.5%Reading at most 65.7
- Vibe Code BenchVals AI79.0%Reading 63.2
- Terminal-Bench 4.0 (AA run)Artificial Analysis22.7%Reading 61.8
- Code MigrationVals AI40.9%Reading 61.1
- Terminal-Bench 4.0Vals AI24.2%Reading 60.5
- SciCodeArtificial Analysis51.3%Reading 53.1
Research and reasoning
4 results · 54.2- ProofBenchVals AI63.0%Reading 62.1
- MysteryMechanismVals AI21.6%Reading 54.3
- Humanity's Last Exam (AA run)Artificial Analysis35.1%Reading 49.9
- Terminal-Bench ScienceVals AINear the floor4.3%Reading at most 49.0
Professional work
7 results · 60.2- Harvey Legal Agent BenchmarkVals AI11.3%Reading 69.7
- Finance AgentVals AI56.3%Reading 66.0
- APEX-AgentsMercor · default effort57.4%Reading 65.8
- EMBVals AI65.5%Reading 63.3
- Tax Agent BenchVals AI28.5%Reading 60.9
- Legal Research BenchVals AI38.0%Reading 59.7
- MedCodeVals AI41.1%Reading 37.6
Knowledge and accuracy
1 result · 46.0- AA-LCRArtificial Analysis74.3%Reading 40.4
Human preference
1 result · 46.4- Arena TextLMArena1,451.8Reading 41.0
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.
- IOIVals AI · CodingReference only47.7%
- SRE BenchVals AI · CodingWatching4.2%
- BioMysteryBenchVals AI · Research and reasoningWatching69.3%
- MedScribeVals AI · Professional workWatching85.3%
- AA Intelligence IndexArtificial Analysis · Composite indicesReference only37.9
- Vals IndexVals AI · Composite indicesReference only53.2%
- Arena VisionLMArena · VisionReference only1,247.5
- Arena WebDevLMArena · Writing and designReference only1,635.7
No result yet
Scored evaluations this model has not taken. A missing result neither adds nor subtracts.
- LiveBench Coding · LiveBench
- Vibe Code Bench 1–100 · 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
- LiveBench Data Analysis · LiveBench
- τ²-Bench Telecom (AA run) · Artificial Analysis
- τ-Bench Banking (AA run) · Artificial Analysis
- SimpleQA Verified · Epoch AI
- LiveBench Language · LiveBench
- LiveBench Instruction Following · LiveBench
- BullshitBench · BullshitBench
- IFBench (AA run) · Artificial Analysis