Tibo 在 2026年8月17日的推文

回复 @kmdrfx

@kmdrfx @thdxr @jayair Dax is obviously the CEO

14 回复 · 3 转推 · 185 赞

What is an obvious thing that we should do with Codex, API or our models that we should just do but haven't yet? What is 100% within reach, but we just seem to be missing?

6,365 回复 · 115 转推 · 5,225 赞

回复 @FintechKristen

@FintechKristen not yet

48 回复 · 2 转推 · 237 赞

回复 @gabrielchua

@gabrielchua @ChatGPT @OpenAIDevs Favorite recent quality feature

47 回复 · 5 转推 · 288 赞

回复 @goldkey

@goldkey Oh no, bummer, another user. Welcome

75 回复 · 7 转推 · 1,796 赞

回复 @ns123abc

@ns123abc I know you’re a fan, you just don’t want to admit it

22 回复 · 1 转推 · 495 赞

回复 @Robertg761_

@Robertg761_ Robert, don’t go

61 回复 · 1 转推 · 388 赞

回复 @Robertg761_

@Robertg761_ Haha

17 回复 · 0 转推 · 104 赞

回复 @ThroughlineRSCH

@ThroughlineRSCH Grock you say. I’ll watch out

31 回复 · 2 转推 · 883 赞

回复 @theisaacmed

@theisaacmed I see from your comments you’ll actually try, nice. Tag me with feedback and I’ll send you some extra credits too

112 回复 · 2 转推 · 1,061 赞

回复 @adamlyttleapps

@adamlyttleapps Welcome

24 回复 · 0 转推 · 164 赞

回复 @ericmitchellai

@ericmitchellai @willdepue Legend

1 回复 · 0 转推 · 40 赞

回复 @AnkanXplorer

@AnkanXplorer Always improving

43 回复 · 2 转推 · 644 赞

回复 @LLMJunky

@LLMJunky No way

24 回复 · 0 转推 · 342 赞

Codex ✅ Almost 100% reliable ✅ Occasional resets ✅ Open-source ✅ (will have Astra) https://t.co/DxNdAgpag5

1,683 回复 · 420 转推 · 13,254 赞

回复 @pwnies

@pwnies Courtesy of Codex https://t.co/z78bvVFGHp

31 回复 · 5 转推 · 646 赞

GPT-5.6 Sol 1M in Codex. This used to only work for API keys, but we just flipped the switch and works for usage through ChatGPT accounts now too. The same warning applies, there is a reason the current context length is the default, we have tuned it to ~perfection. But you do you! Tibo (@thsottiaux) Here is how to enable a 1M-token context window in Codex for GPT-5.6 Sol. Even though we have tuned the context limit in Codex to be set optimally when it comes to performance and cost, this is a common ask, so here it is documented. A larger context window lets Codex retain more code, tool output, and conversation history before summarizing older material. You need a model that supports it. And GPT-5.6 Sol, for example, has a documented 1,050,000-token window. Open ~/.codex/config.toml and add or update these settings at the top level, before any [section] headers: ``` model = "gpt-5.6-sol" model_context_window = 1000000 model_auto_compact_token_limit = 900000 ``` The first setting selects the model. The second tells Codex to use a one-million-token context budget. The third starts automatic history compaction around 900,000 tokens, leaving some headroom. Restart Codex client and start a new session after saving. To try the configuration for a single CLI session without changing your defaults: ``` codex -m gpt-5.6-sol \ -c model_context_window=1000000 \ -c model_auto_compact_token_limit=900000 ``` Have fun, but also know that we tuned the default carefully! — https://x.com/thsottiaux/status/2089082893804896524#m

618 回复 · 253 转推 · 6,057 赞

回复 @ajambrosino

@ajambrosino I had to do a double take

35 回复 · 6 转推 · 1,043 赞