List your company

Efficiency Measurement on Long-form literature review workload

22.4End-to-end Latency (s)Lazy-Attn
21.77625.98830.234.412Jun 3, 2026

Evaluation results

MethodLinks
Lazy-AttnModel=Tulu3-Block-FT 8B, Hardware=H1002026.0622.41.7
Standard RAGModel=Tulu3-Block-FT 8B, Hardware=H1002026.06381