Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Open-ended writing on HelloBench
Loading...
82
Average Score
Qwen3-8B + R2-Write-SFT + RLp
61.2
66.6
72
77.4
Apr 3, 2026
Average Score
Updated 13d ago
Evaluation Results
Method
Method
Links
Average Score
Qwen3-8B + R2-Write-SFT + RLp
Training=SFT + PPO
2026.04
82
LongWriter-Zero-32B
Training=SFT + GRPO
2026.04
80.66
Qwen3-8B + R2-Write-SFT
Training=SFT
2026.04
80.32
Qwen3-4B + R2-Write-SFT + RLp
Training=SFT + PPO
2026.04
79.16
Qwen3-8B + RL
Training=PPO
2026.04
76.8
Qwen3-4B + R2-Write-SFT
Training=SFT
2026.04
76.36
Reverse-Engineering-8B
Training=SFT
2026.04
74.9
Qwen3-8B
Training=-
2026.04
71.8
Qwen3-4B + RL
Training=PPO
2026.04
70.2
Qwen3-4B
Training=-
2026.04
65.86
Longwriter-9B
Training=SFT
2026.04
62
Feedback
Search any
task
Search any
task