Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Long-form generation on LongWriter-Bench
Loading...
94.9
Success Rate (Sl) [0, 500)
Elastic-dLLM
75.556
80.578
85.6
90.622
May 18, 2026
Success Rate (Sl) [0, 500)
Quality Score (Sq) [0, 500)
Success Rate (Sl) [500, 2k)
Quality Score (Sq) [500, 2k)
Success Rate (Sl) [2k, 4k)
Quality Score (Sq) [2k, 4k)
Success Rate (Sl) [4k, 20k)
Quality Score (Sq) [4k, 20k)
Updated 15d ago
Evaluation Results
Method
Method
Links
Success Rate (Sl) [0, 500)
Quality Score (Sq) [0, 500)
Success Rate (Sl) [500, 2k)
Quality Score (Sq) [500, 2k)
Success Rate (Sl) [2k, 4k)
Quality Score (Sq) [2k, 4k)
Success Rate (Sl) [4k, 20k)
Quality Score (Sq) [4k, 20k)
Elastic-dLLM
Model size=1.5, Method...
2026.05
94.9
71.3
92.5
65.6
84.7
46.2
77.4
39.3
Elastic-dLLM
Model size=8B, Method=...
2026.05
86
62.3
91.2
57.3
87.9
42.1
55.6
22.8
LLaDA-8B-Instruct
Model size=8B, Method=...
2026.05
84
48.5
75.7
40.6
51.2
18.6
23.4
4.2
LLaDA-1.5
Model size=1.5, Method...
2026.05
76.3
58.6
85.8
59.9
50.7
34.6
28.7
5.9
Feedback
Search any
task
Search any
task