Claude Sonnet 4.6 (Non-reasoning, High Effort) vs DeepSeek V3.2 Exp (Non-reasoning)
For product management work, compared on the PM Index.
- Claude Sonnet 4.6 (Non-reasoning, High Effort) leads on the PM Index (48.0 vs 29.1).
- DeepSeek V3.2 Exp (Non-reasoning) is cheaper ($0.31 vs $6.00 per 1M tokens).
| Metric | Claude Sonnet 4.6 (Non-reasoning, High Effort) | DeepSeek V3.2 Exp (Non-reasoning) |
|---|---|---|
| PM Index | 48.0 | 29.1 |
| AA Intelligence | 44.4 | 28.4 |
| AA Coding | 46.4 | 30.0 |
| AA Agentic | 61.6 | 31.0 |
| Blended $/1M | $6.00 | $0.31 |
| Output tok/s | 43 | — |