News for hackers. Stuff that matters.
Re:GPT-5.3-Codex by karmasimida on Thursday February 05, 2026 @03:07PM (#46904467)
> With Codex (5.3), the framing is an interactive collaborator: you steer it mid-execution, stay in the loop, course-correct as it works.

> With Opus 4.6, the emphasis is the opposite: a more autonomous, agentic, thoughtful system that plans deeply, runs longer, and asks less of the human.

Ain't the UX is the exact opposite? Codex thinks much longer before gives you back the answer.

Re:GPT-5.3-Codex by WilcoKruijer on Thursday February 05, 2026 @04:11PM (#46905385)
Yes, you’re right for 4.5 and 5.2. Hence they’re focusing on improving the opposite thing and thus are actually converging.
Re:GPT-5.3-Codex by xd1936 on Thursday February 05, 2026 @03:16PM (#46904577)
I've also had the exact opposite experience with tone. Claude Code wants to build with me, and Codex wants to go off on its own for a while before returning with opinions.
Re:GPT-5.3-Codex by mrkstu on Thursday February 05, 2026 @03:37PM (#46904887)
Its likely that both are steering towards the middle from their current relative extremes and converging to nearly the same place.
Re:GPT-5.3-Codex by gervwyk on Thursday February 05, 2026 @04:03PM (#46905277)
also my experience in using these two models. they are trying to recover from oversteer perhaps.
Re:GPT-5.3-Codex by mercnz on Thursday February 05, 2026 @11:02PM (#46908960)
well with the recent delays i can easily find claude code going off on it's own for 20 minutes and have no idea what it's going to come back with. but one time it overflowed it's context on a simple question, and then used up the rest of my session window. in a way a lot of ai assistants have ime have this awkward thing where they complicate something in a non-visible and think about it for a long time burning up context before coming up with a summary based upon some misconception.
Re:GPT-5.3-Codex by esperent on Friday February 06, 2026 @12:07AM (#46909363)
The key is a well defined task with strong guardrails. You can add these to your agents file over time or you can probably just find someone's online to copy the basics from. Any time you find it doing something you didn't expect or don't like, add guardrails to prevent that in future. Claude hooks are also useful here, along with the hookify plugin to create them for you based on the current conversation.
Re:GPT-5.3-Codex by vorticalbox on Friday February 06, 2026 @02:54AM (#46910224)
I have started using openspec for this. I find it works far better to have a proposal and a list of tasks the ai stays more focused.

https://openspec.dev/

Re:GPT-5.3-Codex by zen4ttitude on Saturday February 07, 2026 @02:53AM (#46922123)
For complex tasks I ask ChatGPT or Grok to define context then I take it to Claude for accurate execution. I also created a complete pipeline to use locally and enrich with skills, agents, RAG, profiles. It is slower but very good. There is no magic, the richer the context window the more precise and contained the execution.
Re:GPT-5.3-Codex by kamban on Friday February 06, 2026 @08:49AM (#46912792)
Same here, CC gives me options to pick direction after the planning stage.
Re:GPT-5.3-Codex by PeterStuer on Friday February 06, 2026 @12:00PM (#46915290)
In terms of 'tone', I have been very impressed with Qwen-code-next over the last 2 days, especially as I have it running locally on a single modest 4090.
Re:GPT-5.3-Codex by turtle4 on Friday February 06, 2026 @02:19PM (#46916942)
Did you set that up following a guide or anything you could share?
Re:GPT-5.3-Codex by PeterStuer on Friday February 06, 2026 @03:36PM (#46917813)
Easiest way I know is to just use LMStudio. Just download and press play :). Optional, but recommended, increase the context length to 262144 if you have the DRAM available. It will definitely get slower as your interaction prolongs, but (at least for me) still tolerable speed.
Re:GPT-5.3-Codex by mathrawka on Friday February 06, 2026 @03:00PM (#46917421)
not OP, but I got it running on my 4090 (and RAM) by following this guide: https://unsloth.ai/docs/models/qwen3-coder-next

I see around 30 t/s

Re:GPT-5.3-Codex by bt1a on Thursday February 05, 2026 @05:04PM (#46906040)
This is most likely an inference serving problem in terms of capacity and latency given that Opus X and the latest GPT models available in the API have always responded quickly and slowly, respectively
Re:GPT-5.3-Codex by cwyers on Thursday February 05, 2026 @05:44PM (#46906477)
Codex now lets you tell the LLM tgings in the middle of its thinking without interrupting it, so you can read the thinking traces and tell it to change course if it's going off track.
Re:GPT-5.3-Codex by fluidcruft on Thursday February 05, 2026 @06:33PM (#46906996)
That just seems like a UI difference. I've always interrupted claude code added a comment and it's continued without much issue. Otherwise if you just type the message is queued for next. There's no real reason to prefer one over the other except it sounds like codex can't queue messages?
Re:GPT-5.3-Codex by esperent on Friday February 06, 2026 @12:09AM (#46909370)
[delayed]
Re:GPT-5.3-Codex by cruffle_duffle on Friday February 06, 2026 @12:32AM (#46909477)
That is so annoying too because it basically throws away all the work the subagent did.

Another thing that annoys me is the subagents never output durable findings unless you explicitly tell their parent to prompt the subagent to “write their output to a file for later reuse” (or something like that anyway)

I have no idea how but there needs to be ways to backtrack on context while somehow also maintaining the “future context”…

Re:GPT-5.3-Codex by Skwrm on Friday February 06, 2026 @10:11AM (#46913771)
They actually made a change a few weeks ago that made subagents more steerable

When they ask approval for a tool call, press down til the selector is on "No" and press tab, then you can add any extra instructions

Re:GPT-5.3-Codex by int_19h on Friday February 06, 2026 @04:34AM (#46910812)
Codex can queue messages, but the queue only gets flushed once the agent is done with whatever it was working on, whereas Claude will read messages and adjust accordingly in the middle of whatever it is doing. It sounds like OP is saying that Codex can now do this latter bit as well.