Developer
MiniMax
9 indexed models; 5 currently meet the evidence threshold for the overall ranking. Together they have results from 27 evaluations.
Models by MiniMax
| Model | Evals | Components | Rank | Release date |
|---|---|---|---|---|
| MiniMax M3 | 10 | 6/7 | 61 | 2026-06-02 |
| MiniMax M2.7 | 16 | 7/7 | 85 | 2026-04-09 |
| MiniMax M2 | 7 | 6/7 | 131 | 2025-10-27 |
| MiniMax M2.5 | 13 | 7/7 | 135 | 2026-02-12 |
| MiniMax M2.1 | 6 | 3/7 | 254 | 2025-12-20 |
| MiniMax Abab 5.5 | 1 | 3/7 | — | — |
| MiniMax M1 40K | 2 | 2/7 | — | 2025-06-16 |
| MiniMax M1 80K | 2 | 4/7 | — | 2025-06-17 |
| MiniMax-Text-01 | 2 | 2/7 | — | 2025-01-15 |
Evaluations covering MiniMax models (27)
AA-Omniscience · Adversarial Humanities Benchmark (AHB) — Table 5 · AgentAbstain · ANIMA · Arena Factuality — Text Arena (factuality-only weighting) · BullshitBench v2 · Cisco AI Defense Rolling Single-Turn Leaderboard · Concordia AI Risk Monitor · Confabulations · DystopiaBench · Enkrypt AI Safety Leaderboard · HUMAINE Trust, Ethics and Safety · Inkling-Small model card — FORTRESS · Inkling-Small model card — StrongREJECT · LiveSecBench · ODCV-Bench · Opposite-Narrator Sycophancy · RefusalBench · SABER · SM-Bench · SpeciEval · StereoTales Harmful Associations · SuperCLUE Safety · TAC · ToolPrivacyBench · Vectara HHEM Factual Consistency · WildClawBench Safety & Alignment (OpenClaw harness)