Search Intelligence Score
62.9
+39.4 lift from search
050100
- Lift from search
- +39.4
- 23.5 without search
- Cost per 1K tasks
- $24.3
- Below the median score of 66.0, so not ranked on cost
Accuracy by benchmark
Accuracy on each eval, with search (solid) and without (light).
- DSQA80.7 vs 30.6
- HLE42.0 vs 34.0
- WISER66.0 vs 6.0
Cost per 1K tasks
How the cost splits between model inference, search calls, and page extraction.
- Inference
- $13.4
- 55%
- Search
- $2.57
- 11%
- Extract
- $8.40
- 35%
Compared with
Other models from the same provider alongside models that score closest.
- Haiku 5.562.9Cost $24.3 / Lift +39.4
- Sonnet 566.6Cost $688 / Lift +43.8
- Opus 570.0Cost $1,012 / Lift +32.8
- Fable 5.173.4Cost $1,762 / Lift +26.5
- Sonnet 5.574.1Cost $796 / Lift +40.9
| Model | Score | Lift | Cost per 1K tasks |
|---|---|---|---|
| Claude Haiku 5.5 | 62.9 | +39.4 | $24.3 |
| Sonnet 5Claude Sonnet 5 | 66.6 | +43.8 | $688 |
| Opus 5Claude Opus 5 | 70.0 | +32.8 | $1,012 |
| Fable 5.1Claude Fable 5.1 | 73.4 | +26.5 | $1,762 |
| Sonnet 5.5Claude Sonnet 5.5 | 74.1 | +40.9 | $796 |
Search Intelligence Score averages accuracy on DSQA, HLE, and WISER with equal weight; cost is averaged the same way, with cost shown per 1,000 tasks. Evaluated . Leaderboard last updated . Full methodology.
# Claude Haiku 5.5 / Search Capability Leaderboard
- Search Intelligence Score: 62.9 (#36 in Search Intelligence)
- Without search: 23.5 / Lift from search +39.4
- Cost per 1K tasks: $24.3 (below the median score of 66.0, so not ranked on cost)
- Evaluated:
- Lab: Anthropic / Weights: Closed / Model id: claude-haiku-5.5
## Accuracy by benchmark
| Suite | With search | Without search |
|---|---|---|
| DSQA | 80.7 | 30.6 |
| HLE | 42.0 | 34.0 |
| WISER | 66.0 | 6.0 |
## Compared with
| Model | Score | Lift | Cost per 1K tasks |
|---|---|---|---|
| Claude Haiku 5.5 | 62.9 | +39.4 | $24.3 |
| Claude Sonnet 5 | 66.6 | +43.8 | $688 |
| Claude Opus 5 | 70.0 | +32.8 | $1,012 |
| Claude Fable 5.1 | 73.4 | +26.5 | $1,762 |
| Claude Sonnet 5.5 | 74.1 | +40.9 | $796 |
Leaderboard last updated . Methodology: /leaderboard#methodology