2026-09-28 · America/Los_Angeles · 社区动态 · #15

ImaJev-4b: I spent 15 days fine-tuning a 4B model to make business decisions from text and photos, and it just ranked #1 of 91 on JevBench & ahead of GPT-5.6 Luna on DecisionBench

Some context first. I am process improvement / business consultant and had worked with Fortune 500 companies on improving their processes around refunds, returns, customer support, etc. This entire thing had a lot of complex decision making and generally every decision / condition node in a process map was generally replaced by a human because factoring in ambiguity in a code is very difficult. Idea of ImaJev Hence, when Jev came out, I was very intrigued with it and also could clearly see its use-case of improving decision making in complex decision work flows. However, Jev didnt have…

热度 46.6 / 100；排名与评分保留该期记录。