Search Intelligence Score
64.2
+33.6 lift from search
050100
- Lift from search
- +33.6
- 30.6 without search
- Cost per 1K tasks
- $266
- #8 in Search Efficiency
- Time per task
- 432s
- Search plus inference time to complete one task
Accuracy by benchmark
Accuracy on each eval, with search (solid) and without (light).
- DSQA86.5 vs 44.7
- HLE45.0 vs 35.0
- WISER61.0 vs 12.0
Cost per 1K tasks
How the cost splits between model inference, search calls, and page extraction.
- Inference
- $247
- 93%
- Search
- $6.43
- 2%
- Extract
- $12.2
- 5%
Compared with
Other models from the same provider alongside models that score closest.
- Kimi K364.2Cost $266 · Lift +33.6 · 432s
- Gemini 3.865.1Cost $177 · Lift +26.1 · 317s
- DS Flash 4.162.7Cost $35.8 · Lift +35.7 · 549s
- GLM 5.362.7Cost $78.5 · Lift +40.8 · 308s
- Muse Spark 1.366.1Cost $142 · Lift +35.8 · 274s
| Model | Score | Lift | Cost per 1K tasks | Time per task |
|---|---|---|---|---|
| Kimi K3 | 64.2 | +33.6 | $266 | 432s |
| Gemini 3.8Gemini 3.8 Flash | 65.1 | +26.1 | $177 | 317s |
| DS Flash 4.1DeepSeek V4.1 Flash | 62.7 | +35.7 | $35.8 | 549s |
| GLM 5.3GLM 5.3 | 62.7 | +40.8 | $78.5 | 308s |
| Muse Spark 1.3Muse Spark 1.3 | 66.1 | +35.8 | $142 | 274s |
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.
# Kimi K3 · Search Capability Leaderboard
- Search Intelligence Score: 64.2 (#11 in Search Intelligence)
- Without search: 30.6 · Lift from search +33.6
- Cost per 1K tasks: $266 (#8 in Search Efficiency)
- Time per task: 432s
- Lab: Moonshot AI · Model id: moonshotai/kimi-k3
## Accuracy by benchmark
| Suite | With search | Without search |
|---|---|---|
| DSQA | 86.5 | 44.7 |
| HLE | 45.0 | 35.0 |
| WISER | 61.0 | 12.0 |
## Compared with
| Model | Score | Lift | Cost per 1K tasks | Time per task |
|---|---|---|---|---|
| Kimi K3 | 64.2 | +33.6 | $266 | 432s |
| Gemini 3.8 Flash | 65.1 | +26.1 | $177 | 317s |
| DeepSeek V4.1 Flash | 62.7 | +35.7 | $35.8 | 549s |
| GLM 5.3 | 62.7 | +40.8 | $78.5 | 308s |
| Muse Spark 1.3 | 66.1 | +35.8 | $142 | 274s |
Latest update September 25, 2026. Methodology: /leaderboard#methodology