Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Deep Research on GAIA Text-Only
Loading...
80.1
Score
REDSearcher-30B-A3B
36.212
47.606
59
70.394
Apr 21, 2026
Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Score
REDSearcher-30B-A3B
Model Category=Trained...
2026.04
80.1
GPT-5 High
Model Category=Foundat...
2026.04
76.4
SMTL-30B-300
Model Category=Trained...
2026.04
75.7
DeepSeek-V3.2
Model Category=Foundat...
2026.04
75.1
WebSailor-V2-30B-RL
Model Category=Trained...
2026.04
74.1
Tongyi-DR-30B
Model Category=Trained...
2026.04
70.9
DR-Venus-4B-SFT
Model Category=Trained...
2026.04
65.4
DR-Venus-4B-RL
Model Category=Trained...
2026.04
64.4
MiniMax-M2.1
Model Category=Foundat...
2026.04
64.3
OpenResearcher-30B-A3B
Model Category=Trained...
2026.04
64.1
AgentCPM-Explore-4B
Model Category=Trained...
2026.04
63.9
GLM-4.7
Model Category=Foundat...
2026.04
61.9
DeepMiner-32B-RL
Model Category=Trained...
2026.04
58.7
OffSeeker-8B-DPO
Model Category=Trained...
2026.04
51.5
WebExplorer-8B-RL
Model Category=Trained...
2026.04
50
OffSeeker-8B-SFT
Model Category=Trained...
2026.04
47.6
WebSailor-7B
Model Category=Trained...
2026.04
37.9
Feedback
Search any
task
Search any
task