[djoldman]

For day-to-day programming work, have you seen a difference in the quality of output between (Opus 4.6 / GPT 5.2 / GPT-5.3 Codex) and the current (GPT-5.6 / Fable) that justifies the price increase?

My intuition says that the output quality difference is marginal compared to the change in price especially when taking into account the effects of prompt/context engineering and harness differences.

Essentially: since opus 4.6, working through a model's quirks with prompt/context engineering and harness development will yield significantly better output than just switching models to the latest.

[edot]

I’m starting to come to the opposite approach: don’t try to customize anything, just use it vanilla, and use the best model you can afford. No AGENTS.md, no special subagents or roles, nothing but a few convenience skills which are really just textexpander. Use the harness that the LLM provider makes, and that’s it.

Making a huge custom setup is so 2025.

[amunozo]

I agree with this. However, aren't harnesses like Claude Code a bit bloated? Would it be better to use something like Pi?

[edot]

Yeah, but the bloat is “correct” per the manufacturer. I view it like car parts or other things where the OEM (original equipment manufacturer) recommends certain things. Besides, they have the most training data and incentive to get their harness working as well as possible with their LLM.

[wronex]

Every time I try one of the newer models I don’t want to go back. What is the value of a dumper model? It makes more mistakes. Wastes more of my time. At anything below Opus 4.8 I’m better of writing code myself. As a tools, it needs to outperform me. Unfortunately, it tends to be lazy. Which is rather ironic from a machine. We taught it well. Alignment is not an issue :’)