Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Agentic Task Performance on ACEBench Agent (test)
Loading...
56.7
Average Score
CURATEEVO
27.06
34.755
42.45
50.145
Jul 7, 2026
Average Score
Updated 17d ago
Evaluation Results
Method
Method
Links
Average Score
CURATEEVO
Model=QWEN3-4B, Data S...
2026.07
56.7
CURATEEVO
Model=QWEN3-4B, Data S...
2026.07
55.8
MUA-RL
Model=QWEN3-8B, Data S...
2026.07
53.3
MUA-RL
Model=QWEN3-8B, Data S...
2026.07
50.7
EnvScaler
Model=QWEN3-8B, Data S...
2026.07
50.4
EnvScaler
Model=QWEN3-8B, Data S...
2026.07
48.8
RODS
Model=QWEN3-4B, Data S...
2026.07
48.3
AWM
Model=QWEN3-8B, Data S...
2026.07
47.8
RODS
Model=QWEN3-4B, Data S...
2026.07
46.2
AWM
Model=QWEN3-8B, Data S...
2026.07
45.9
FunReason-MT
Model=QWEN3-4B, Data S...
2026.07
45.7
FunReason-MT
Model=QWEN3-4B, Data S...
2026.07
43.4
GRPO w/o. Data Curation
Model=QWEN3-8B, Data S...
2026.07
36.8
GRPO w/o. Data Curation
Model=QWEN3-4B, Data S...
2026.07
32.1
GRPO w/o. Data Curation
Model=QWEN3-8B, Data S...
2026.07
30.9
GRPO w/o. Data Curation
Model=QWEN3-4B, Data S...
2026.07
28.2
Feedback
Search any
task
Search any
task