Share your thoughts, 1 month free Claude Pro on usSee more

Science on GPQA D

69.9Accuracy

Length Scaling

Updated 2mo ago

Evaluation Results

Method	Links
Length Scaling 2026.04		69.9
Length Scaling 2026.04		68.2
Reasoning Memory 2026.04		68.1
Document RAG (Google) 2026.04		67.8
Reasoning Memory 2026.04		67.4
No RAG 2026.04		66.9
Template RAG (ReasonFlux) 2026.04		64.4
Template RAG (Self-Discover) 2026.04		63.3
Trajectory RAG (Prefix) 2026.04		60.7
Document RAG (CompactDS) 2026.04		60.2
Trajectory RAG (Summary) 2026.04		59.5
Reasoning Memory 2026.04		54.2
Reasoning Memory 2026.04		52.2
Length Scaling 2026.04		50.2
Length Scaling 2026.04		48.4
Reasoning Memory 2026.04		46.1
Document RAG (CompactDS) 2026.04		45.8
Template RAG (ReasonFlux) 2026.04		45.7
Document RAG (Google) 2026.04		45.2
Trajectory RAG (Summary) 2026.04		45.1
Length Scaling 2026.04		44.7
Reasoning Memory 2026.04		44.4
Template RAG (Self-Discover) 2026.04		44.1
No RAG 2026.04		43.8
Trajectory RAG (Prefix) 2026.04		43.7
Trajectory RAG (Summary) 2026.04		43.1
No RAG 2026.04		42.7
Length Scaling 2026.04		42.6
Template RAG (ReasonFlux) 2026.04		41.7
Template RAG (Self-Discover) 2026.04		41.4
Trajectory RAG (Prefix) 2026.04		39.1
Document RAG (Google) 2026.04		38.3
Document RAG (CompactDS) 2026.04		31.9