GPT-5.5 (xhigh) vs Grok 4.20 0309 (Reasoning)
For product management work, compared on the PM Index.
- GPT-5.5 (xhigh) leads on the PM Index (62.9 vs 48.4).
- Grok 4.20 0309 (Reasoning) is cheaper ($3.00 vs $11.25 per 1M tokens).
- Grok 4.20 0309 (Reasoning) is faster (197 vs 64 tok/s).
| Metric | GPT-5.5 (xhigh) | Grok 4.20 0309 (Reasoning) |
|---|---|---|
| PM Index | 62.9 | 48.4 |
| AA Intelligence | 60.2 | 48.5 |
| AA Coding | 59.1 | 42.2 |
| AA Agentic | 74.1 | 50.9 |
| Blended $/1M | $11.25 | $3.00 |
| Output tok/s | 64 | 197 |