TSCG 20 Tools: leaderboard
Metric: Scenario D, 20 tools, TSCG compilation: tool-selection accuracy (%) (self-reported). Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 3 12B | 95 |
| 2 | phi-4-14B | 84.4 |
| 3 | Qwen 3 14B | 84.1 |
| 4 | Llama 3.1 8B | 81 |
| 5 | Mistral 7B | 80.1 |
| 6 | Qwen 3 4B | 69.3 |
| 7 | Gemma 3 4B | 67 |
Interactive version: theaggregate.ai/benchmark?slug=tscg-20-tools · How It Works · Data refreshed daily, snapshot 2026-10-09.