I don’t believe this benchmark 27b size model next opus 4.5! Anyone can confirm testing with real agentic workflow?

r/LocalLLaMA
Generative AI AI Research

Submitted by /u/Wonderful-Ad-5952 [link] [comments]