tencent
Comparing 2 models from tencent.
| Model | Total ▼ | Released | Context | CoT | Tooling | Creative Writing | Language | Utility | Reasoning | Text Editing | Rule Following | Hallucination |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Hy4 Preview (Reasoning, High) | 94.80% | Aug 28, 26 | 1m | ✓ | 100.00% | 87.32% | 97.11% | 99.90% | 89.85% | 98.60% | 91.44% | 94.17% |
| Hy4 (Reasoning, High) | 90.20% | Nov 4, 26 | 1m | ✓ | 100.00% | 90.18% | 100.00% | 94.11% | 96.60% | 97.45% | 56.45% | 86.83% |
Model Performance
Cost vs Performance
Compares total benchmark cost against overall score for tencent models. Quadrant lines are drawn at the median values.
Cost Breakdown
Total benchmark cost per model, broken down by input, reasoning, and output tokens. Toggle between USD and token views.