Search Intelligence Score
61.1
+33.3 lift from search
050100
- Lift from search
- +33.3
- 27.8 without search
- Cost per 1K tasks
- $118
- 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).
- DSQA78.3 vs 35.3
- HLE43.0 vs 39.0
- WISER62.0 vs 9.0
Cost per 1K tasks
How the cost splits between model inference, search calls, and page extraction.
- Inference
- $99.9
- 85%
- Search
- $5.03
- 4%
- Extract
- $13.0
- 11%
Compared with
Other models from the same provider alongside models that score closest.
- Step 5 Preview61.1Cost $118 / Lift +33.3
- Ultramarine Flash61.1Cost $87.2 / Lift +37.1
- MiMo v2.6 Flash60.9Cost $25.2 / Lift +38.5
- GLM 5.3 Flash60.8Cost $16.6 / Lift +40.4
- GPT-5.6 Luna60.7Cost $36.2 / Lift +33.9
| Model | Score | Lift | Cost per 1K tasks |
|---|---|---|---|
| Step 5 Preview | 61.1 | +33.3 | $118 |
| Ultramarine FlashUltramarine 1 Flash | 61.1 | +37.1 | $87.2 |
| MiMo v2.6 FlashMiMo v2.6 Flash | 60.9 | +38.5 | $25.2 |
| GLM 5.3 FlashGLM 5.3 Flash | 60.8 | +40.4 | $16.6 |
| GPT-5.6 LunaGPT-5.6 Luna | 60.7 | +33.9 | $36.2 |
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.
# Step 5 Preview / Search Capability Leaderboard
- Search Intelligence Score: 61.1 (#40 in Search Intelligence)
- Without search: 27.8 / Lift from search +33.3
- Cost per 1K tasks: $118 (below the median score of 66.0, so not ranked on cost)
- Evaluated:
- Lab: StepFun / Weights: Open weights / Model id: stepfun/step-5-preview
## Accuracy by benchmark
| Suite | With search | Without search |
|---|---|---|
| DSQA | 78.3 | 35.3 |
| HLE | 43.0 | 39.0 |
| WISER | 62.0 | 9.0 |
## Compared with
| Model | Score | Lift | Cost per 1K tasks |
|---|---|---|---|
| Step 5 Preview | 61.1 | +33.3 | $118 |
| Ultramarine 1 Flash | 61.1 | +37.1 | $87.2 |
| MiMo v2.6 Flash | 60.9 | +38.5 | $25.2 |
| GLM 5.3 Flash | 60.8 | +40.4 | $16.6 |
| GPT-5.6 Luna | 60.7 | +33.9 | $36.2 |
Leaderboard last updated . Methodology: /leaderboard#methodology