Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Instruction Following on IFBench (Pr. (S))
Loading...
57.3
Pr. (S)
Hybrid Reward
11.644
23.497
35.35
47.203
May 28, 2026
Pr. (S)
Updated 5d ago
Evaluation Results
Method
Method
Links
Pr. (S)
Hybrid Reward
Backbone=GLM-4.7-Flash
2026.05
57.3
GLM-4.7-Flash
RL Training=Baseline
2026.05
50.8
Hybrid Reward
Backbone=Qwen3-30B-A3B
2026.05
39.3
Hybrid Reward
Backbone=Qwen3-4B
2026.05
35.9
Qwen3-30B-A3B
RL Training=Baseline
2026.05
35.9
Qwen3-4B
RL Training=Baseline
2026.05
29.4
Hybrid Reward
Backbone=DeepSeek-R1-D...
2026.05
21.3
DeepSeek-R1-Distill-Qwen-7B
RL Training=Baseline
2026.05
13.4
Feedback
Search any
task
Search any
task