Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
High-level Reasoning on HLE (Text, MM, All)
Loading...
29.7
HLE Score (Text)
ReAct
22.212
24.156
26.1
28.044
Jun 30, 2026
HLE Score (Text)
HLE Score (MM)
HLE Score (All)
Updated 24d ago
Evaluation Results
Method
Method
Links
HLE Score (Text)
HLE Score (MM)
HLE Score (All)
ReAct
Main LLM=Gemini-3.1-fl...
2026.06
29.7
26.5
29
ReAct + ACE
Main LLM=Gemini-3.1-fl...
2026.06
29.5
29.2
29.4
ReAct + Truncation
Main LLM=Gemini-3.1-fl...
2026.06
29.2
28.3
29
ReAct + Summarization
Main LLM=Gemini-3.1-fl...
2026.06
25.8
26.5
26
ReAct + ACE
Main LLM=GPT-4.1, Fram...
2026.06
24.8
22.1
24.2
ReAct + Truncation
Main LLM=GPT-4.1, Fram...
2026.06
23.5
23
23.4
ReAct + Summarization
Main LLM=GPT-4.1, Fram...
2026.06
23
23
23
ReAct
Main LLM=GPT-4.1, Fram...
2026.06
22.5
21.2
22.2
Feedback
Search any
task
Search any
task