claude-opus-5 vs claude-sonnet-5
Each agent graded on the same private suites, running headless through its own tools. Advantages are oriented so a win is a win, whichever way the metric runs.
| Dimension | claude-opus-5 | claude-sonnet-5 | Advantage |
|---|---|---|---|
| Capability | 57% | 53% | +3% (claude-opus-5 leads) |
| Median task time | 3m 57s | 5m 6s | 1m 9s faster (claude-opus-5 leads) |
| Approx. task cost | $0.010 / task | $0.009 / task | $0.001 / task lower (claude-sonnet-5 leads) |
claude-opus-5: 11.0% of Claude Max 20x (11.0% / 5h) per run. claude-sonnet-5: about 9.7% of Claude Max 20x (about 9.7% / 5h, 1.0% / 1w) per run (estimated: 2 of 15 cells measured no share of their own). Each figure is a share of that agent's own plan, so there is no delta to draw.
Work per window: claude-opus-5 1328 vs claude-sonnet-5 1503 suite runs per month (claude-sonnet-5 leads); per plan-dollar: 6.64 vs 7.52 (claude-sonnet-5 leads). Work figures compare outputs, so unlike quota shares they are comparable across plans.
| Dimension | claude-opus-5 | claude-sonnet-5 | Advantage |
|---|---|---|---|
| Capability | 54% | 37% | +17% (claude-opus-5 leads) |
| Median task time | 6m 57s | 5m 21s | 1m 36s faster (claude-sonnet-5 leads) |
| Approx. task cost | $0.016 / task | $0.005 / task | $0.011 / task lower (claude-sonnet-5 leads) |
claude-opus-5: 1.2% of Claude Max 20x (1.2% / 5h) per good solve. claude-sonnet-5: 0.4% of Claude Max 20x (0.4% / 5h, 0.2% / 1w) per good solve. Each figure is a share of that agent's own plan, so there is no delta to draw.
Work per window: claude-opus-5 12175 vs claude-sonnet-5 36525 good solves per month (claude-sonnet-5 leads); per plan-dollar: 60.88 vs 182.63 (claude-sonnet-5 leads). Work figures compare outputs, so unlike quota shares they are comparable across plans.
Grades and deltas are comparable only within a suite version. See methodology.
Frequently asked questions
Which is better for coding, claude-opus-5 or claude-sonnet-5?
claude-sonnet-5 clears higher: tier 3 of 5 versus tier 2 for claude-opus-5 on real-world@v3, our private suite of real-world coding problems. Latest measurement 2026-08-10.
Which gives more work per dollar, claude-opus-5 or claude-sonnet-5?
claude-sonnet-5 delivers more benchmark work per subscription dollar: 7.52 suite runs per plan-dollar versus 6.64 for claude-opus-5 (at $200/month priced 2026-08-05, and $200/month priced 2026-08-05). Work figures compare outputs, so unlike raw quota shares they are comparable across plans.
How much do claude-opus-5 and claude-sonnet-5 cost?
claude-opus-5 runs on Claude Max 20x at $200/month (public price as of 2026-08-05). claude-sonnet-5 runs on Claude Max 20x at $200/month (public price as of 2026-08-05).