MiniMax-Text-01: benchmark results
MiniMax's open 456B MoE (45.9B active) with hybrid Lightning Attention and up to a 4M-token inference context (January 2025). Provider: MiniMax. Released 2025-01-14. Access: Open.
Aggregate Rating 1579 ± 12, rank #513 of 1656 rated models, from 232 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| pfgen-bench - Chat Mode - Truthfulness | 0.99 | Truthfulness Score | 100 |
| pfgen-bench - QA Mode - Truthfulness | 0.99 | Truthfulness Score | 100 |
| OpenCompass Code - Code QA - Chinese (CompassBench 2501) | 80.6 | Score (%) | 97.4 |
| OpenCompass Language - Dialogue - English (CompassBench 2501) | 72.6 | Score (%) | 97.4 |
| OpenCompass Language - NLP - Chinese (CompassBench 2501) | 71.3 | Score (%) | 97.4 |
| pfgen-bench - Chat Mode - Fluency | 0.86 | Fluency Score | 97.1 |
| pfgen-bench - Chat Mode - Helpfulness | 0.64 | Helpfulness Score | 97.1 |
| pfgen-bench - Chat Mode - Score | 0.83 | pfgen Score (mean of three) | 97.1 |
| pfgen-bench - QA Mode - Score | 0.82 | pfgen Score (mean of three) | 95.2 |
| OpenCompass Code - Code QA - English (CompassBench 2501) | 69.4 | Score (%) | 94.9 |
| OpenCompass LLM - Instruction Following - English (CompassBench 2501) | 67.3 | Score (%) | 94.9 |
| pfgen-bench - QA Mode - Helpfulness | 0.61 | Helpfulness Score | 94.5 |
Interactive version: theaggregate.ai/model?slug=minimax-text-01 · How It Works · Data refreshed daily, snapshot 2026-10-20.