GPT-5.2 vs MiniMax-M3
For product management work, compared on the PM Index.
- MiniMax-M3 leads on the PM Index (49.1 vs 46.5).
- MiniMax-M3 is cheaper ($0.52 vs $4.81 per 1M tokens).
- GPT-5.2 is faster (77 vs 60 tok/s).
| Metric | GPT-5.2 | MiniMax-M3 |
|---|---|---|
| PM Index | 46.5 | 49.1 |
| AA Intelligence | 42.2 | 44.4 |
| AA Coding | 48.7 | 43.4 |
| AA Agentic | 60.2 | 68.6 |
| Blended $/1M | $4.81 | $0.52 |
| Output tok/s | 77 | 60 |