Search Intelligence Score
58.6
+24.5 lift from search
050100
- Lift from search
- +24.5
- 34.1 without search
- Cost per 1K tasks
- $114
- Below the median score of 61.9, so not ranked on cost
- Time per task
- 276s
- Search plus inference time to complete one task
Accuracy by benchmark
Accuracy on each eval, with search (solid) and without (light).
- DSQA82.8 vs 54.4
- HLE39.0 vs 33.0
- WISER54.0 vs 15.0
Cost per 1K tasks
How the cost splits between model inference, search calls, and page extraction.
- Inference
- $103
- 90%
- Search
- $9.24
- 8%
- Extract
- $1.88
- 2%
Compared with
Other models from the same provider alongside models that score closest.
- Gemini 358.6Cost $114 · Lift +24.5 · 276s
- Gemini 3.865.1Cost $177 · Lift +26.1 · 317s
- Gemini 3.766.8Cost $130 · Lift +27.4 · 229s
- DS Pro58.4Cost $110 · Lift +27.9 · 383s
- Hunyuan 358.2Cost $28.4 · Lift +35.6 · 363s
| Model | Score | Lift | Cost per 1K tasks | Time per task |
|---|---|---|---|---|
| Gemini 3 Flash | 58.6 | +24.5 | $114 | 276s |
| Gemini 3.8Gemini 3.8 Flash | 65.1 | +26.1 | $177 | 317s |
| Gemini 3.7Gemini 3.7 Flash | 66.8 | +27.4 | $130 | 229s |
| DS ProDeepSeek V4 Pro | 58.4 | +27.9 | $110 | 383s |
| Hunyuan 3Hunyuan 3 | 58.2 | +35.6 | $28.4 | 363s |
Search Intelligence Score averages accuracy on DSQA, HLE, and WISER with equal weight; cost and time are averaged the same way, with cost shown per 1,000 tasks. Latest update September 25, 2026. Full methodology.
# Gemini 3 Flash · Search Capability Leaderboard
- Search Intelligence Score: 58.6 (#17 in Search Intelligence)
- Without search: 34.1 · Lift from search +24.5
- Cost per 1K tasks: $114 (below the median score of 61.9, so not ranked on cost)
- Time per task: 276s
- Lab: Google · Model id: google/gemini-3-flash-preview
## Accuracy by benchmark
| Suite | With search | Without search |
|---|---|---|
| DSQA | 82.8 | 54.4 |
| HLE | 39.0 | 33.0 |
| WISER | 54.0 | 15.0 |
## Compared with
| Model | Score | Lift | Cost per 1K tasks | Time per task |
|---|---|---|---|---|
| Gemini 3 Flash | 58.6 | +24.5 | $114 | 276s |
| Gemini 3.8 Flash | 65.1 | +26.1 | $177 | 317s |
| Gemini 3.7 Flash | 66.8 | +27.4 | $130 | 229s |
| DeepSeek V4 Pro | 58.4 | +27.9 | $110 | 383s |
| Hunyuan 3 | 58.2 | +35.6 | $28.4 | 363s |
Latest update September 25, 2026. Methodology: /leaderboard#methodology