Qwen3-30B-A3B-Instruct-2507-FP8: benchmark results
Provider: Alibaba. Released 2025-07-29. Access: Open.
Aggregate Rating 1552 ± 50, rank #1685 of 10916 rated models, from 36 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Prompt Tool Sensitivity - Graphs Reasoning | 0.92 | Accuracy (0–1) | 100 |
| Prompt Tool Sensitivity - Mathematics Direct Answer | 0.46 | Accuracy (0–1) | 100 |
| Prompt Tool Sensitivity - Molecules Tools | 1 | Accuracy (0–1) | 100 |
| Prompt Tool Sensitivity - Music Constant Tools | 0.48 | Accuracy (0–1) | 100 |
| Prompt Tool Sensitivity - Music First-Token Read | 0.43 | Accuracy (0–1) | 100 |
| Prompt Tool Sensitivity - Music Fixed Readout | 0.45 | Accuracy (0–1) | 100 |
| Prompt Tool Sensitivity - Music Tools | 0.6 | Accuracy (0–1) | 100 |
| Prompt Tool Sensitivity - Tables Constant Tools | 0.68 | Accuracy (0–1) | 100 |
| Prompt Tool Sensitivity - Tables Tools | 0.7 | Accuracy (0–1) | 100 |
| Prompt Tool Sensitivity - Graphs Constant Tools | 0.83 | Accuracy (0–1) | 90 |
| Prompt Tool Sensitivity - Graphs Supplied Report | 0.98 | Accuracy (0–1) | 90 |
| Prompt Tool Sensitivity - Molecules Supplied Report | 1 | Accuracy (0–1) | 90 |
Interactive version: theaggregate.ai/model?slug=qwen3-30b-a3b-instruct-2507-fp8 · How It Works · Data refreshed daily, snapshot 2026-11-08.