Hunyuan 3 Preview: benchmark results
Provider: Tencent.
Aggregate Rating 1715 ± 44, rank #283 of 10916 rated models, from 15 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| GPAgentBench-2K - Diagnosis Accuracy | 66.4 | Diagnosis accuracy (%; final diagnosis judged correct by a G | 90 |
| GPAgentBench-2K - Missed Referral Rate | 46.8 | Missed referral rate (%; referral-required cases the agent m | 90 |
| GPAgentBench-2K - Treatment Score | 40.7 | Treatment score (%; prescribed treatment scored by a GPT-5.4 | 70 |
| ProductWebBench - Vision Task Success | 2.34 | Rate (%) | 62.5 |
| ProductWebBench - Available-task Average | 2.34 | Rate (%) | 58.3 |
| ProductWebBench - Language Task Success | 2.34 | Rate (%) | 58.3 |
| GPAgentBench-2K - Management Accuracy | 53 | Management accuracy (%; episodes ending in the correct pathw | 47.5 |
| GPAgentBench-2K - Missed Local Management Rate | 47.2 | Missed local management rate (%; locally manageable cases th | 30 |
| AutoLab | 31 | Overall Score (imported) | 20 |
| ProductWebBench - Consistency | 47.72 | Rate (%) | 16.7 |
| ProductWebBench - Edit | 52.15 | Rate (%) | 16.7 |
| ProductWebBench - Stage Success | 34.69 | Rate (%) | 16.7 |
Interactive version: theaggregate.ai/model?slug=hunyuan-3-preview · How It Works · Data refreshed daily, snapshot 2026-11-08.