Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Information Retrieval and Deep Research on xbench DeepSearch 2505
Loading...
83.7
Score
Step-3.5-Flash
65.812
70.456
75.1
79.744
Jun 8, 2026
Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Score
Step-3.5-Flash
Size=196B-A11B, Contex...
2026.06
83.7
SearchSwarm
Size=30B-A3B, Context...
2026.06
80.8
DeepSeek V3.2
Size=671B-A37B, Contex...
2026.06
78
LongSeeker
Size=30B-A3B, Context...
2026.06
78
GPT-5
Size=-, Context manage...
2026.06
77.8
Tongyi DeepResearch
Size=30B-A3B, Context...
2026.06
75
Tongyi DR Swarm
Size=30B-A3B, Context...
2026.06
75
MiroThinker-1.5-mini
Size=30B-A3B, Context...
2026.06
73.1
GLM-4.7
Size=355B-A32B, Contex...
2026.06
72
MiniMax-M2
Size=230B-A10B, Contex...
2026.06
72
Claude-4.5-Sonnet
Size=-, Context manage...
2026.06
66.5
Feedback
Search any
task
Search any
task