Creative Writing on Blessing short-text
94.6ScoreDoubao-Seed-1.6
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Doubao-Seed-1.62026.04 | 94.6 | — | |
| Qwen3-8B-Thinking + RLBase Model=Qwen3-8B, Mode=Thinking, RL Optimization=UniCreative2026.04 | 93.6 | 25.6 | |
| Claude-Sonnet-4.52026.04 | 93.2 | — | |
| Claude-Sonnet-42026.04 | 92.8 | — | |
| Qwen3-4B-Thinking + RLBase Model=Qwen3-4B, Mode=Thinking, RL Optimization=UniCreative2026.04 | 91.4 | 17.4 | |
| DeepSeek-V3.22026.04 | 90.4 | — | |
| Qwen3-1.7B-Thinking + RLBase Model=Qwen3-1.7B, Mode=Thinking, RL Optimization=UniCreative2026.04 | 90 | 25.8 | |
| GPT-4.12026.04 | 89.4 | — | |
| Qwen3-32B-ThinkingMode=Thinking2026.04 | 88.4 | — | |
| DeepSeek-V3-2503242026.04 | 87.2 | — | |
| Grok-42026.04 | 80.6 | — | |
| Qwen3-4B-ThinkingBase Model=Qwen3-4B, Mode=Thinking2026.04 | 74 | — | |
| Qwen3-8B-ThinkingBase Model=Qwen3-8B, Mode=Thinking2026.04 | 68 | — | |
| Qwen3-1.7B-ThinkingBase Model=Qwen3-1.7B, Mode=Thinking2026.04 | 64.2 | — | |
| Qwen3-32B2026.04 | 45.2 | — | |
| Qwen3-8BBase Model=Qwen3-8B2026.04 | 43.6 | — | |
| Qwen3-4BBase Model=Qwen3-4B2026.04 | 40.8 | — | |
| Qwen3-1.7BBase Model=Qwen3-1.7B2026.04 | 8.4 | — |