Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Instruction Following on Instruction Following Suite IFB, IFE
Loading...
27.3
IFB Score
CARE-RL
14.508
17.829
21.15
24.471
May 30, 2026
IFB Score
IFE Score
Updated 1d ago
Evaluation Results
Method
Method
Links
IFB Score
IFE Score
CARE-RL
Backbone=Qwen3-4B
2026.05
27.3
48.9
CARE-RL
Backbone=Qwen2.5-7B
2026.05
27.2
41.8
MOPD
Backbone=Qwen3-4B
2026.05
26.9
47.6
NV→V
Backbone=Qwen3-4B
2026.05
26.8
46.6
MGS
Backbone=Qwen3-4B
2026.05
26.3
46.9
MOPD
Backbone=Qwen2.5-7B
2026.05
26.1
41.2
Naive Mixing
Backbone=Qwen3-4B
2026.05
26
45.7
V→NV
Backbone=Qwen3-4B
2026.05
25.6
45
MGS
Backbone=Qwen2.5-7B
2026.05
25.3
40.4
V→NV
Backbone=Qwen2.5-7B
2026.05
24.8
38.2
NV→V
Backbone=Qwen2.5-7B
2026.05
24.4
39.2
Naive Mixing
Backbone=Qwen2.5-7B
2026.05
24.2
38.9
Base
Backbone=Qwen3-4B
2026.05
18
39.4
Base
Backbone=Qwen2.5-7B
2026.05
15
31.2
Feedback
Search any
task
Search any
task