MiniMax
4 ranked models in a catalog of 9 text models. MiniMax-M3 leads this lab at #23 overall.
Ranked models
Global ranks · current snapshot| Rank | Model | SI Score | coding | math | reasoning | preference | Confidence | Blended price | Context | Released |
|---|---|---|---|---|---|---|---|---|---|---|
| #23 | MiniMax-M3Open weights | 68.0 | 54.6 | 70.7 | 80.3 | 76.2 | 100% | $0.52 | 1M | Jun 1, 2026 |
| #70 | MiniMax-M2.7Open weights | 57.9 | 62.4 | — | — | 73.6 | 88% | $0.52 | 205K | Mar 18, 2026 |
| #72 | MiniMax-M2.5Open weights | 57.8 | 75.8 | — | 33.3 | 68.9 | 100% | $0.52 | 205K | Feb 12, 2026 |
| #84 | MiniMax-M2Open weights | 55.7 | 64.5 | — | — | 66.8 | 96% | $0.52 | 205K | Oct 27, 2025 |
Pillars use a 0–100 scale. * marks an imputed neutral prior where the model has no scored results in that pillar; it is not a measured benchmark result. Prices are USD per 1M tokens; see the blend and price comparison.
Scores and releases
Provisional models
5 awaiting rank-eligible evidence- MiniMax-M3.1-Flash-Preview
No usable open benchmark results found
- MiniMax-M2.7-highspeed
No usable open benchmark results found
- MiniMax-M2.5-highspeed
No usable open benchmark results found
- MiniMax-M2 Her
No usable open benchmark results found
- MiniMax-M2.1
Results cover 1 pillar; at least 2 required