| ▲ | jubilanti a day ago |
| The fact there is no way to disable auto-compaction and no way to go back in the conversation history to before a compact makes codex a no-go for me on any codebase > 5kloc. Even worse it seems to fire randomly at 10-20% of context left. So really you have 80% of 272k as real usable context. Compaction kills my sessions, it hallucinates and is worse than starting fresh. I've had enough times screaming at my computer when it burns tokens on a large codebase, gets to 15%, auto-compacts, and hallucinates so bad it has to read the entire codebase agin, gets to 15%, auto-compacts.... |
|
| ▲ | volker48 a day ago | parent | next [-] |
| Have you tried using another harness like Pi? You can easily turn off auto compaction and if you leave it on you can use /tree to go back before the compaction if you like. |
| |
| ▲ | sdesol a day ago | parent [-] | | > you can use /tree to go back before the compaction This is really a killer feature in my opinion. I'm currently working on pi brains extension that is designed to solve the compaction issue I have, which is, I don't know what the AI knows after compaction. https://gitsense.com/screenshots/pi-inspect-overview.png The screenshot shot above contains what I call a compaction capsule which contains all the messages/events at the point of compaction. The "Files" section in the right side panel shows me how many files were read/modified since compaction. After compacting, I can easily tell what Pi use to know and use `/tree` to go back in time. I still wouldn't compact more than 5 times before starting a new session since compaction messages persists in Pi. | | |
|
|
| ▲ | mycall 19 hours ago | parent | prev | next [-] |
| So the practical workaround is to set model_auto_compact_token_limit very near, but still below, the model’s context window. This postpones compaction rather than disabling it. config.toml model_context_window = 27000
model_auto_compact_token_limit = 17000
model_auto_compact_token_limit_scope = "total"
|
|
| ▲ | behnamoh 18 hours ago | parent | prev | next [-] |
| > The fact there is no way to disable auto-compaction and no way to go back in the conversation history to before a compact makes codex a no-go for me Same. I've told @tibo many times about these issues but apparently OpenAI's focus is on features that buy them the most users ASAP instead of increasing harness quality. |
|
| ▲ | p1esk a day ago | parent | prev [-] |
| [flagged] |
| |
| ▲ | jubilanti a day ago | parent | next [-] | | I did find a strategy, thank you very much, and it was easy: npm uninstall -g @openai/codex && curl -fsSL https://claude.ai/install.sh | bash
This was the final straw that got me to downgrade my OpenAI subscription and move the bulk of my spending to Anthropic and Claude code.OpenAI devs, if you're reading, this is how you lost a customer. | | |
| ▲ | no_no_no_yes a day ago | parent | next [-] | | Claude does the same? It asks you to create new chats (Claude Design) and Claude Code nudges you to start over too, I think `/compact` is optional though? | | |
| ▲ | jwitthuhn 21 hours ago | parent [-] | | Claude does the same when you hit the context limit, but the difference is that claude's context limit is 1m and Codex's is 272k now. Claude lets you get a lot more done before you have to worry about the start getting compacted away. | | |
| ▲ | anukin 19 hours ago | parent | next [-] | | Have you observed any decrease in performance at context higher than 200k? That used to be one of the major issues why clearing or compacting context when the context nears 150 to 170k is usually preferred. | | |
| ▲ | ilc 17 hours ago | parent | next [-] | | Opus: Yeah, it can be a little fuzzy at 600-700k, I prefer it 400K and under in general. But, I've had usable sessions to 850k. It really depends on the session, and what you are doing how far you can push it. | |
| ▲ | PeterStuer 10 hours ago | parent | prev | next [-] | | I see it more past 300-350k these days, but probably there is a smaller decline even past 250k. | |
| ▲ | jwitthuhn 19 hours ago | parent | prev [-] | | Not really, on the project I'm working on right now most of my longer prompts seems to consume somewhere between 300k and 400k tokens (input + output) and I find that both Opus 4.8 and Fable 5 don't seem to get lost with that much context. At least the results still look good to me. I don't think I've ever pushed past 500k though, it may well fall apart at the higher end of what is officially supported. |
| |
| ▲ | Jimmc414 15 hours ago | parent | prev | next [-] | | There are numerous Claude settings to disable or modify this behavior | |
| ▲ | phonon 19 hours ago | parent | prev [-] | | Codex context limit is 400k total, including 128,000 reserved output tokens. |
|
| |
| ▲ | youre-wrong3 14 hours ago | parent | prev | next [-] | | So you went from a bad product to a worse product? There’s good solutions out there and you’re like “nope don’t want a good solution. I want a worse solution.” ^ this is literally what the op did. Don’t flag facts. | |
| ▲ | throw1234567891 18 hours ago | parent | prev | next [-] | | [dead] | |
| ▲ | youre-wrong3 19 hours ago | parent | prev [-] | | [flagged] |
| |
| ▲ | Kye a day ago | parent | prev [-] | | This would have been a great opportunity to suggest one. | | |
|
