MiniMax·released 2026-04-09sources disagree
MiniMax-M2.7 benchmark scores: 5 benchmarks tracked. 0% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 5 benchmarks·best result 45.1 on Terminal Bench (unverified)·0 of 5 independently reproduced
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| Terminal Bench | 45.1accuracy (%) | unverified· optimizedT1 | 2026-03-18https://www.tbench.ai/leaderboard/terminal-bench/2.0 |
| WeirdML | 36.95accuracy (%) | unverifiedT1 | 2026-03-18https://htihle.github.io/weirdml.html |
| ExploitBench | 13.3accuracy (%) | unverifiedT2 | 2026-03-18 |
| ProofBench | 3accuracy (%) | unverifiedT1 | 2026-03-18https://www.vals.ai/benchmarks/proof_bench |
| CritPt | 0.57accuracy (%) | unverifiedT2 | 2026-03-18 |
Claims drawn from cited facts, not live model generation.
Developed by MiniMax, this model was released in 2026. With 228.69 billion parameters, this is a frontier-scale model, offering ample headroom but at correspondingly high computational cost.
3 cited facts
This model is tracked on 5 benchmarks and currently holds no top score on any of them. The scores are unverified: outside evaluation has neither confirmed nor contradicted them, so treat the results as claims rather than verified facts.
3 cited facts
This model trails the state-of-the-art leader by an average of 51.28 points, a wide gap that places it well behind the frontier. It holds no top scores on the benchmark set (leader count 0), so it shows no genuine category leadership.
2 cited facts
6 facts cross-checked across data sources: 5 single-source, 1 disputed.
Source facts, citations, and refresh stamp for this record.
Sources: epoch_benchmarks, huggingface_models, openrouter_models
MiniMax-M2.7 is an AI model developed by MiniMax, released 2026-04-09. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
MiniMax-M2.7 has recorded scores on 5 benchmarks, each shown with its evidence status.
MiniMax-M2.7 has recorded scores on 5 benchmarks — WeirdML, CritPt, ProofBench, Terminal Bench, ExploitBench. The full table above shows each score with its evidence status.
0 of 5 recorded scores (0%) are independently reproduced rather than self-reported by the lab.
MiniMax-M2.7 has 5 tracked claims: 5 unverified.