I still want something no one has, which is the ability to launch agents in different git worktrees simultaneously and check the results out on my main branch for testing when they are finished.
Would this be similar to how Charlie and Jules work?
Building more with GPT-5.1-Codex-Max
11–20 of 332 posts
Re: Building more with GPT-5.1-Codex-Max
#12I still want something no one has, which is the ability to launch agents in different git worktrees simultaneously and check the results out on my main branch for testing when they are finished.
I think I’ve described how I achieve kinda your desired workflow in a comment yesterday [0].
Re: Building more with GPT-5.1-Codex-Max
#13> Compaction enables GPT‑5.1-Codex-Max to complete tasks that would have previously failed due to context-window limits, such as complex refactors and long-running agent loops by pruning its history while preserving the most important context over long horizons. In Codex applications, GPT‑5.1-Codex-Max automatically compacts its session when it approaches its context window limit, giving it a fresh context window. It repeats this process until the task is completed.
Wouldn't the model automatically do that using attention techniques? Why do you need to do it at the token layer and not leave it to the model to automatically decide which tokens are worth paying attention to?
Re: Building more with GPT-5.1-Codex-Max
#14These 2 sentences right next to each other stood out to me: > a new step towards becoming a reliable coding partner > GPT‑5.1-Codex-Max is built for long-running, detailed work Does this not sound contradictory? It’s been the shorter form work that has built what little confidence I have in these as a coding partner - a model that goes off and does work without supervision is not a partner to me.
Absolutely contradictory. The long-running tendency for Codex is why I cannot understand the hype around it: if you bother to watch what it does and read its code the approaches it takes are absolutely horrifying. It would rather rewrite a TLS library from scratch than bother to ask you if the network is available.
Re: Building more with GPT-5.1-Codex-Max
#15Rest assured that we are better at training models than naming them ;D
- New benchmark SOTAs with 77.9% on SWE-Bench-Verified, 79.9% on SWE-Lancer, and 58.1% on TerminalBench 2.0
- Natively trained to work across many hours across multiple context windows via compaction
- 30% more token-efficient at the same reasoning level across many tasks
Let us know what you think!
Re: Building more with GPT-5.1-Codex-Max
#16> Compaction enables GPT‑5.1-Codex-Max to complete tasks that would have previously failed due to context-window limits, such as complex refactors and long-running agent loops by pruning its history while preserving the most important context over long horizons. In Codex applications, GPT‑5.1-Codex-Max automatically compacts its session when it approaches its context window limit, giving it a fresh context window. It…
> due to context-window limits
Re: Building more with GPT-5.1-Codex-Max
#17Sigh. Time to try it again I guess. I give OpenAI way more chances than it deserves.
Re: Building more with GPT-5.1-Codex-Max
#18Gemini 3 had a great 24 hour SOTA run for coding
Re: Building more with GPT-5.1-Codex-Max
#19The new detergent now washes even whiter
Re: Building more with GPT-5.1-Codex-Max
#20Rest assured that we are better at training models than naming them ;D - New benchmark SOTAs with 77.9% on SWE-Bench-Verified, 79.9% on SWE-Lancer, and 58.1% on TerminalBench 2.0 - Natively trained to work across many hours across multiple context windows via compaction - 30% more token-efficient at the same reasoning level across many tasks Let us know what you think!
did you address this https://github.com/openai/codex/issues/6426 ?
how much more token efficient is this compared to 5.0
had to use 5.0 because 5.1 was eating tokens like crazy and seemed like a slight incremental improvement barely noticeable