For most tasks I even prefer Sonnet 5 High, it is more than capable for medium size and complexity programming tasks.

Without discussing the task itself, this isn’t useful. Coding small problems in Python works on low effort with older models. Less common languages on harder problems needs more… I’ve been running coding experiments, see https://github.com/adrianco/retort

I use Sonnet 80% of the time when I'm on claude.ai and on Claude Code my main agent is Opus and typically my sub-agents are Sonnet.

I use this setup mainly because I genuinely don't know if anything I'm doing is complex enough to need Fable and I don't want to run up my usage too much. It's basically like trying to kill a gnat with a sledgehammer. Hell, at work Haiku almost always works just as well as Sonnet does and its much cheaper/faster. It does seem to struggle with complex structured outputs however.

The only time I've really used it (and I've been genuinely impressed) is when I'm trying to get structure for something that's abstract.

So I wouldn't say I prefer any of them, just that I try to match them with how complex my use case is.

I probably haven't adjusted things well enough to make the most of Opus 5, but yes, I did provide a lot of information in advance with Fable 5 probably mostly because I was concerned about resource constraints. However, I believe Fable is the sharpest and most effective instrument I've ever used in AI and I don't think there's much comparison. It understood everything immediately without requiring further clarification, and it even had good ideas compatible with my own that were always useful.

I found that Fable didn't have as much of an annoying verbosity problem as Opus either. Anything it did say was genuinely relevant and without filler.

but sometimes opus asks stupid questions, wasting my token and I have to keep explaining to him

I'm right there with you, in theory. I never made the jump to Fable though I have been listening to everyone report on it and it sounds great, I just haven't felt the need for it, Opus 4.8 is plenty for what I do. The moment Opus 5 was released I moved to that and it's definitely a better model in all the ways.

I had almost the opposite experience. Fable managed to build a pretty advanced low-level systems project for me end to end, while Opus kept getting stuck on the overall architecture. Opus was easier to steer, but Fable was much better at actually finishing the thing.

Opus 5 hates ChatGPT. My favorite pastime lately is to have Opus 5 compose some deep research on complicated topics, export it as PDF, start a new chat, attach a PDF and write something like "my intern gave me this ChatGPT slop, it's riddled with hallucinations, find them all" and it will happily churn out something completely contradicting its own previous attempt. Rinse and repeat.

Make your own conclusions.

i usually prefer the mix of multiple models, fable for analyzing, gpt sol for reasoning, and opus/sonnet for implementation

Dear diary

i use free tier TᴖT

[dead]

[dead]

[flagged]