2026-02-23 · 1 min
how agentic coding effects LLMs
airesearch
A comprehensive analysis of agentic capabilities across eight leading models from different labs
Notes / Reflections
A comprehensive analysis of agentic capabilities across eight leading models from different labs
Four new models just dropped this week this is an quick review of all of them
qwen3.5 while being 1/3 and 1/2 the size of best open-weight models it still managed to be in top 3 of open-weight models just behind glm-5