Audio2Tool (Text Oracle) - Parametric Commands: leaderboard
Metric: Exact match (%): share of queries whose predicted tool or ordered tool sequence and all arguments exactly match the reference after deterministic normalization, on the 5,800 Tier 2 parametric commands (slot filling with constrained arguments) of Audio2Tool, with the ground-truth text of the query given to a text LLM (the oracle setting that removes ASR errors), zero-shot; higher is better. Source: arxiv.org. Saturation forecast: Around June 2027. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 3 12B | 12.3 |
| 2 | Qwen 3 8B | 10.1 |
| 3 | Qwen 3 4B | 6.7 |
| 4 | Qwen 3 1.7B | 3.2 |
Interactive version: theaggregate.ai/benchmark?slug=audio2tool-text-oracle-parametric-commands · How It Works · Data refreshed daily, snapshot 2026-10-07.