ESF-Bench - Selection: leaderboard
Metric: Tagged slot accuracy (%; slots tagged with a SEL (Selection) scenario, choosing among candidate values from the conversation, sources or schema enums; 810 synthetic multi-turn enterprise slot-filling samples (6,530 slots, 8 domains) whose prompts combine a schema, a user-assistant conversation and auxiliary sources; varied output formats mapped to a standard form before exact comparison with the ground truth; 32k maximum tokens, reasoning setting as printed per model). Source: arxiv.org. Saturation forecast: Around December 2026. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Flash | 89.5 |
| 2 | Qwen 3 32B (Thinking) | 83.1 |
| 3 | GPT-5.1 (High) | 82.6 |
| 4 | GPT-OSS-120B (Low) | 81.5 |
| 5 | Ministral-3-14B-Reasoning-2512 | 80.8 |
| 6 | GPT-OSS-120B (High) | 80.3 |
| 7 | GPT-5.1 (Non-reasoning) | 75.6 |
| 8 | Gemini 2.5 Flash (Non-reasoning) | 72.3 |
| 9 | Ministral-3-14B-Instruct-2512 | 68.6 |
| 10 | Qwen 3 32B (Non-reasoning) | 64.8 |
Interactive version: theaggregate.ai/benchmark?slug=esf-bench-selection · How It Works · Data refreshed daily, snapshot 2026-09-29.