DataSense-Bench - BFCL Data Selection: leaderboard
Metric: Selected-data BFCL accuracy (%). Source: datasense-bench.github.io. Saturation forecast: Around 2036. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V4 Pro | 36.53 | #96 |
| 2 | GPT-6 Astra | 36.49 | #6 |
| 3 | Kimi K3 | 36.29 | #25 |
| 4 | Claude Opus 5 | 35.94 | #11 |
| 5 | Claude Fable 5.1 | 35.21 | #8 |
| 6 | Gemini 3.5 Flash | 35.1 | #56 |
No result here: #3 Claude Opus 5.5, #9 GPT-6.1 Sol, #10 Claude Sonnet 5.5.
Interactive version: theaggregate.ai/benchmark?slug=datasense-bench-bfcl-data-selection · How It Works · Data refreshed daily, snapshot 2026-10-12.