Search Intelligence Score
66.0
+44.8 lift from search
050100
- Lift from search
- +44.8
- 21.1 without search
- Cost per 1K tasks
- $252
- 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).
- DSQA79.0 vs 18.4
- HLE49.0 vs 36.0
- WISER70.0 vs 9.0
Cost per 1K tasks
How the cost splits between model inference, search calls, and page extraction.
- Inference
- $235
- 94%
- Search
- $7.22
- 3%
- Extract
- $8.99
- 4%
Compared with
Other models from the same provider alongside models that score closest.
- Grok 4.766.0Cost $252 / Lift +44.8
- Qwen3.8 Max66.0Cost $221 / Lift +37.1
- Muse Spark 1.366.1Cost $142 / Lift +35.8
- Sonnet 566.6Cost $688 / Lift +43.8
- GPT-6 Sol66.6Cost $183 / Lift +28.7
| Model | Score | Lift | Cost per 1K tasks |
|---|---|---|---|
| Grok 4.7 Medium | 66.0 | +44.8 | $252 |
| Qwen3.8 MaxQwen3.8 Max | 66.0 | +37.1 | $221 |
| Muse Spark 1.3Muse Spark 1.3 | 66.1 | +35.8 | $142 |
| Sonnet 5Claude Sonnet 5 | 66.6 | +43.8 | $688 |
| GPT-6 SolGPT-6 Sol | 66.6 | +28.7 | $183 |
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.
# Grok 4.7 Medium / Search Capability Leaderboard
- Search Intelligence Score: 66.0 (#29 in Search Intelligence)
- Without search: 21.1 / Lift from search +44.8
- Cost per 1K tasks: $252 (below the median score of 66.0, so not ranked on cost)
- Evaluated:
- Lab: xAI / Weights: Closed / Model id: x-ai/grok-4.7
## Accuracy by benchmark
| Suite | With search | Without search |
|---|---|---|
| DSQA | 79.0 | 18.4 |
| HLE | 49.0 | 36.0 |
| WISER | 70.0 | 9.0 |
## Compared with
| Model | Score | Lift | Cost per 1K tasks |
|---|---|---|---|
| Grok 4.7 Medium | 66.0 | +44.8 | $252 |
| Qwen3.8 Max | 66.0 | +37.1 | $221 |
| Muse Spark 1.3 | 66.1 | +35.8 | $142 |
| Claude Sonnet 5 | 66.6 | +43.8 | $688 |
| GPT-6 Sol | 66.6 | +28.7 | $183 |
Leaderboard last updated . Methodology: /leaderboard#methodology