List your company

Science on GPQA D

69.9AccuracyLength Scaling
30.3840.6450.961.16Apr 1, 2026

Evaluation results

MethodLinks
Length ScalingBackbone=Qwen3-32B, Ret.=N/A, m=82026.0469.9
Length ScalingBackbone=Qwen3-32B, Ret.=N/A, m=302026.0468.2
Reasoning MemoryBackbone=Qwen3-32B, Ret.=Proc., m=302026.0468.1
Document RAG (Google)Backbone=Qwen3-32B, Ret.=Fact., m=82026.0467.8
Reasoning MemoryBackbone=Qwen3-32B, Ret.=Proc., m=82026.0467.4
No RAGBackbone=Qwen3-32B, Ret.=N/A, m=82026.0466.9
Template RAG (ReasonFlux)Backbone=Qwen3-32B, Ret.=Proc., m=82026.0464.4
Template RAG (Self-Discover)Backbone=Qwen3-32B, Ret.=Proc., m=82026.0463.3
Trajectory RAG (Prefix)Backbone=Qwen3-32B, Ret.=Proc., m=82026.0460.7
Document RAG (CompactDS)Backbone=Qwen3-32B, Ret.=Fact., m=82026.0460.2
Trajectory RAG (Summary)Backbone=Qwen3-32B, Ret.=Proc., m=82026.0459.5
Reasoning MemoryBackbone=OpenThinker3-7B, Ret.=Proc., m=302026.0454.2
Reasoning MemoryBackbone=OpenThinker3-7B, Ret.=Proc., m=82026.0452.2
Length ScalingBackbone=OpenThinker3-7B, Ret.=N/A, m=302026.0450.2
Length ScalingBackbone=OpenThinker3-7B, Ret.=N/A, m=82026.0448.4
Reasoning MemoryBackbone=DeepSeek-R1-Distill-Llama-8B, Ret.=Proc., m=302026.0446.1
Document RAG (CompactDS)Backbone=OpenThinker3-7B, Ret.=Fact., m=82026.0445.8
Template RAG (ReasonFlux)Backbone=OpenThinker3-7B, Ret.=Proc., m=82026.0445.7
Document RAG (Google)Backbone=OpenThinker3-7B, Ret.=Fact., m=82026.0445.2
Trajectory RAG (Summary)Backbone=OpenThinker3-7B, Ret.=Proc., m=82026.0445.1
Length ScalingBackbone=DeepSeek-R1-Distill-Llama-8B, Ret.=N/A, m=302026.0444.7
Reasoning MemoryBackbone=DeepSeek-R1-Distill-Llama-8B, Ret.=Proc., m=82026.0444.4
Template RAG (Self-Discover)Backbone=OpenThinker3-7B, Ret.=Proc., m=82026.0444.1
No RAGBackbone=OpenThinker3-7B, Ret.=N/A, m=82026.0443.8
Trajectory RAG (Prefix)Backbone=OpenThinker3-7B, Ret.=Proc., m=82026.0443.7
Trajectory RAG (Summary)Backbone=DeepSeek-R1-Distill-Llama-8B, Ret.=Proc., m=82026.0443.1
No RAGBackbone=DeepSeek-R1-Distill-Llama-8B, Ret.=N/A, m=82026.0442.7
Length ScalingBackbone=DeepSeek-R1-Distill-Llama-8B, Ret.=N/A, m=82026.0442.6
Template RAG (ReasonFlux)Backbone=DeepSeek-R1-Distill-Llama-8B, Ret.=Proc., m=82026.0441.7
Template RAG (Self-Discover)Backbone=DeepSeek-R1-Distill-Llama-8B, Ret.=Proc., m=82026.0441.4
Trajectory RAG (Prefix)Backbone=DeepSeek-R1-Distill-Llama-8B, Ret.=Proc., m=82026.0439.1
Document RAG (Google)Backbone=DeepSeek-R1-Distill-Llama-8B, Ret.=Fact., m=82026.0438.3
Document RAG (CompactDS)Backbone=DeepSeek-R1-Distill-Llama-8B, Ret.=Fact., m=82026.0431.9