Claude Haiku 4.5 vs Claude Sonnet 5
Claude Sonnet 5 is no longer on the leaderboard. The grades below are the last ones measured, so read them as history rather than as this week's board. See the current roster.
Claude Sonnet 5 leads overall, D to F. Claude Haiku 4.5 is about 2.0x cheaper per task. By suite, Claude Sonnet 5 takes real-world issues (D+ to D-) and spec planning (D to F). Take Claude Haiku 4.5 anyway when you want the side that is cheaper per task and faster on real-world issues.Measured
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.
Verdict: Claude Sonnet 5 vs Claude Haiku 4.5
Claude Sonnet 5 leads overall, D to F. By suite: Claude Sonnet 5 takes real-world issues, D+ to D- (53% vs 37% capability); Claude Sonnet 5 takes spec planning, D to F (44% vs 21% capability); vibe coding is even at F (22% vs 31% capability). Claude Haiku 4.5 used 1% of Claude Max 5x's weekly limit per pass, $0.012 per task at the listed price. Claude Sonnet 5 used 2% of Claude Max 5x's weekly limit per pass, $0.024 per task at the listed price. Per task, Claude Haiku 4.5 is about 2.0x cheaper ($0.012 vs $0.024). Claude Haiku 4.5 is faster on real-world issues: a median task takes 3m 8s against 3m 35s. Both finished every task on it. Reasons to pick Claude Haiku 4.5 anyway: it is cheaper per task ($0.012 vs $0.024) and it is faster on real-world issues (3m 8s vs 3m 35s median).
Changed since W39: Claude Haiku 4.5: real-world issues D+ to D-; Claude Sonnet 5: vibe coding D+ to F.
| Dimension | Claude Haiku 4.5 | Claude Sonnet 5 | Advantage |
|---|---|---|---|
| Capability | 37% | 53% | +17% (Claude Sonnet 5 leads) |
| Median task time | 3m 8s | 3m 35s | 27s faster (Claude Haiku 4.5 leads) |
| Approx. task cost | $0.015 / task | $0.015 / task | Even |
Claude Haiku 4.5: 8.0% of Claude Max 5x (8.0% / 5h, 1.0% / 1w) per run. Its weekly limit holds about 13.8 of the week's 33.6 5h windows. Claude Sonnet 5: 12.0% of Claude Max 5x (12.0% / 5h, 1.0% / 1w) per run. Its weekly limit holds about 13.8 of the week's 33.6 5h windows. Each figure is a share of that agent's own plan, so there is no delta to draw.
Work per window: Claude Haiku 4.5 1826 vs Claude Sonnet 5 1218 suite runs per month (Claude Haiku 4.5 leads); per plan-dollar: 18.26 vs 12.18 (Claude Haiku 4.5 leads). Work figures compare outputs, so unlike quota shares they are comparable across plans.
| Dimension | Claude Haiku 4.5 | Claude Sonnet 5 | Advantage |
|---|---|---|---|
| Capability | 21% | 44% | +23% (Claude Sonnet 5 leads) |
| Median task time | 2m 8s | 5m 12s | 3m 4s faster (Claude Haiku 4.5 leads) |
| Approx. task cost | $0.002 / task | $0.057 / task | $0.056 / task lower (Claude Haiku 4.5 leads) |
Claude Haiku 4.5: 1.0% of Claude Max 5x (1.0% / 5h) per run. Its weekly limit holds about 13.8 of the week's 33.6 5h windows. Claude Sonnet 5: 5.0% of Claude Max 5x (5.0% / 5h, 1.0% / 1w) per run. Its weekly limit holds about 13.8 of the week's 33.6 5h windows. Each figure is a share of that agent's own plan, so there is no delta to draw.
Work per window: Claude Haiku 4.5 14610 vs Claude Sonnet 5 2922 suite runs per month (Claude Haiku 4.5 leads); per plan-dollar: 146.10 vs 29.22 (Claude Haiku 4.5 leads). Work figures compare outputs, so unlike quota shares they are comparable across plans.
| Dimension | Claude Haiku 4.5 | Claude Sonnet 5 | Advantage |
|---|---|---|---|
| Capability | 22% | 31% | +9% (Claude Sonnet 5 leads) |
| Median task time | 4m 4s | 4m 31s | 27s faster (Claude Haiku 4.5 leads) |
| Approx. task cost | $0.009 / task | $0.007 / task | $0.002 / task lower (Claude Sonnet 5 leads) |
Claude Haiku 4.5: 1.3% of Claude Max 5x (1.3% / 5h) per good solve. Its weekly limit holds about 13.8 of the week's 33.6 5h windows. Claude Sonnet 5: 1.0% of Claude Max 5x (1.0% / 5h) per good solve. Its weekly limit holds about 13.8 of the week's 33.6 5h windows. Each figure is a share of that agent's own plan, so there is no delta to draw.
Work per window: Claude Haiku 4.5 11688 vs Claude Sonnet 5 14610 good solves per month (Claude Sonnet 5 leads); per plan-dollar: 116.88 vs 146.10 (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 Haiku 4.5 or Claude Sonnet 5?
Claude Sonnet 5 clears higher: tier 3 of 5 versus tier 2 for Claude Haiku 4.5 on real-world@v3, our private suite of real-world coding problems. Latest measurement 2026-09-28.
Which is cheaper per task, Claude Haiku 4.5 or Claude Sonnet 5?
Claude Haiku 4.5 used 1% of Claude Max 5x's weekly limit per pass, $0.012 per task at the listed price. Claude Sonnet 5 used 2% of Claude Max 5x's weekly limit per pass, $0.024 per task at the listed price. Per task, Claude Haiku 4.5 is about 2.0x cheaper ($0.012 vs $0.024).
Which gives more coding work per dollar, Claude Haiku 4.5 or Claude Sonnet 5?
Claude Haiku 4.5 delivers more benchmark work per subscription dollar: 18.26 suite runs per plan-dollar versus 12.18 for Claude Sonnet 5 (at $100/month priced 2026-07-20, and $100/month priced 2026-07-20). Work figures compare outputs, so unlike raw quota shares they are comparable across plans.
What do Claude Haiku 4.5 and Claude Sonnet 5 cost a month?
Claude Haiku 4.5 runs on Claude Max 5x at $100/month (public price as of 2026-07-20). Claude Sonnet 5 runs on Claude Max 5x at $100/month (public price as of 2026-07-20).