B.O.B. Leaderboard
MiniMaxreleased 12 Feb 2026
MiniMax-M2.5
#86 of 9512 of 46 benchmarksHolds up
11.2
B.O.B.
Category profile
- Real use (hard to game)2×14
- Reasoning & knowledge1
- Math9
- Coding15
- Agents & real work0
- Independent indexes25
Reality check
Lab tests
6
Real use (6 tests)
14
Does even better in real use than on lab tests. It does 8 points better on tests that are hard to train for than on the usual lab tests.
What counts as real usePrice per 1M tokens (OpenRouter)
$0.27 in · $0.95 out
Cost per task on the AA index
—
Output speed (Artificial Analysis)
—
§Receipts
Every benchmark result
“Beats” = the share of tracked models this one beat on that test. The raw score is its best reported setting.
| Benchmark | Raw | Beats | Source |
|---|---|---|---|
| Real use (hard to game) · Real use 14 | |||
| Kagi LLM Benchmarkas minimax-m2-5 | 55.2% | 33%of 22 | Kagi |
| SWE-rebenchas MiniMax M2.5 | 47.7% | 30%of 34 | Nebius |
| LMArena WebDevas minimax-m2.5 | 1386 | 17%of 82 | LMArena (blind human votes) |
| LMArena Textas minimax-m2.5 | 1391 | 3%of 80 | LMArena (blind human votes) |
| EQ-Bench Creative Writingas minimax/minimax-m2.5 | 1361 | 2%of 46 | Sam Paech (open source) |
| SkateBenchas minimax-m2.5 | 14.0% | 0%of 8 | Theo (t3.gg, YouTube) |
| Reasoning & knowledge · Reasoning 1 | |||
| ARC-AGI-2as MiniMax-M2.5 | 4.9% | 1%of 44 | Epoch AI Benchmarking Hub |
| Math · Math 9 | |||
| ProofBenchas MiniMax-M2.5 | 4.0% | 9%of 68 | Epoch AI Benchmarking Hub |
| Coding · Coding 15 | |||
| Terminal-Benchas MiniMax-M2.5 | 42.7% | 15%of 14 | Epoch AI Benchmarking Hub |
| ALE-Benchas MiniMax-M2.5 | 618.2% | 14%of 71 | Epoch AI Benchmarking Hub |
| Agents & real work · Agents 0 | |||
| Vending-Bench 2as MiniMax-M2.5 | -2316.0% | 0%of 47 | Epoch AI Benchmarking Hub |
| Independent indexes · Indexes 25 | |||
| Epoch Capabilities Indexas MiniMax-M2.5 | 146.7 | 25%of 81 | Epoch AI |