Creative Writing on HelloBench
80.4AccuracyQwen3-30B-A3B-Thinking
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen3-30B-A3B-ThinkingThinking Mode=Thinking2026.04 | 80.4 | 0.7 | |
| Qwen3-30B-A3B-InstructThinking Mode=No Thinking2026.04 | 79.8 | — | |
| DeepSeek-R1-0528Thinking Mode=Thinking2026.04 | 78 | 0.7 | |
| DeepSeek-V3.1Thinking Mode=No Thinking2026.04 | 77.4 | — | |
| Qwen3-8BThinking Mode=Thinking2026.04 | 72.2 | 0.4 | |
| Qwen3-8BThinking Mode=No Thinking2026.04 | 71.9 | — |