Search Intelligence Score
65.3
+34.3 lift from search
050100
- Lift from search
- +34.3
- 31.0 without search
- Cost per 1K tasks
- $98.2
- #4 in Search Efficiency
- Time per task
- 66.2s
- Search plus inference time to complete one task
Accuracy by benchmark
Accuracy on each eval, with search (solid) and without (light).
- DSQA76.0 vs 24.0
- HLE59.0 vs 55.0
- WISER61.0 vs 14.0
Cost per 1K tasks
How the cost splits between model inference, search calls, and page extraction.
- Inference
- $91.8
- 94%
- Search
- $4.08
- 4%
- Extract
- $2.29
- 2%
InferenceSearchExtract
Compared with
Other models from the same provider alongside models that score closest.
| Model | Score | Lift | Cost per 1K tasks | Time per task |
|---|---|---|---|---|
| Muse Spark 1.3 | 65.3 | +34.3 | $98.2 | 66.2s |
| DS Flash 4.1DeepSeek V4.1 Flash | 65.3 | +45.0 | $44.8 | 124s |
| GPT-5.6 SolGPT-5.6 Sol | 66.7 | +35.3 | $217 | 52.9s |
| Fable 5.1Claude Fable 5.1 | 61.0 | +29.1 | $455 | 81.5s |
| GPT-6 (Astra)GPT-6 (Astra) | 70.3 | +35.3 | $301 | 34.7s |
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 14, 2026. Full methodology.
# Muse Spark 1.3 · Search Capability Leaderboard
- Search Intelligence Score: 65.3 (#4 in Search Intelligence)
- Without search: 31.0 · Lift from search +34.3
- Cost per 1K tasks: $98.2 (#4 in Search Efficiency)
- Time per task: 66.2s
- Lab: Meta · Model id: meta/muse-spark-1.3
## Accuracy by benchmark
| Suite | With search | Without search |
|---|---|---|
| DSQA | 76.0 | 24.0 |
| HLE | 59.0 | 55.0 |
| WISER | 61.0 | 14.0 |
## Compared with
| Model | Score | Lift | Cost per 1K tasks | Time per task |
|---|---|---|---|---|
| Muse Spark 1.3 | 65.3 | +34.3 | $98.2 | 66.2s |
| DeepSeek V4.1 Flash | 65.3 | +45.0 | $44.8 | 124s |
| GPT-5.6 Sol | 66.7 | +35.3 | $217 | 52.9s |
| Claude Fable 5.1 | 61.0 | +29.1 | $455 | 81.5s |
| GPT-6 (Astra) | 70.3 | +35.3 | $301 | 34.7s |
Latest update September 14, 2026. Methodology: /leaderboard#methodology