Many users prefer Sol over Fable for Hermes agent tasks because it is more thorough, persistent, and reliable with fewer refusals or drifting.
Based on 15 visible X reactions from 72 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@Teknium I've been using GPT 5.5 in Hermes the past few months. It worked well but not perfect. Now, with GPT 5.6 Sol, it works really well. Actually, it works great. It would create subagents and continue to work when I thought it was done. It even double checked its work. I love it.
@Teknium GPT 5.6 Sol is a lot less target drifting, do task very persistently and sticks tight to the plan, asked high quality verification questions without you asking. It’s very pleasant to use
@Teknium I’m a big fan of Sol so far. I had 5.5 building an app for the last 2 weeks and Sol found critical issues on the first pass. Incredible how well it is working so far
Early days using Sol for Hermes Agent dev, and so far.. I think I like it more than Fable. I've been using claude models for at least the last year primarily, but I just tried it on a relatively common heavy task - and Fable simply refused me of course, but, Sol not only didn't refuse but felt way more thorough, and its obviously way cheaper, and faster. Will be probably trying Sol out for the rest of the week to see how I feel about it across a lot more dev tasks in Hermes Agent.
Many users prefer Sol over Fable for Hermes agent tasks because it is more thorough, persistent, and reliable with fewer refusals or drifting.
Based on 15 visible X reactions from 72 accounts; directional sample.
Ask a question below.
Published answers will appear here.