TSCG 50 Tools json-text: leaderboard
Metric: Scenario D, 50 tools, json-text baseline: tool-selection accuracy (%) (self-reported). Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 14B | 94.6 |
| 2 | Qwen 3 4B | 90 |
| 3 | Gemma 3 12B | 85 |
| 4 | Llama 3.1 8B | 75.1 |
| 5 | Mistral 7B | 30 |
| 6 | Gemma 3 4B | 24.3 |
| 7 | phi-4-14B | 0 |
Interactive version: theaggregate.ai/benchmark?slug=tscg-50-tools-json-text · How It Works · Data refreshed daily, snapshot 2026-10-09.