Adrian @agentx_Diff

LA Joined June 2016

Tweets

66
Followers

17
Following

51
Likes

21

Adrian @agentx_Diff

2 days ago

@Lonely__MH @thsottiaux 20x so fast

1 0 0 554 0

View Details

Most devs still think Codex = OpenAI models only. That framing is outdated. Tibo just reminded people that Codex App, CLI, and SDK can run with open-source models too, not just OpenAI models. This changes the game: Codex is becoming less of a “single-model product” and more of an agent runtime. Swap the model. Keep the workflow. Run OSS / local / self-hosted. Optimize for cost, privacy, latency, or control. The next wave of coding agents may not be about “which model wins.” It may be about who owns the best agent layer on top of any model. Which open-source model would you plug into Codex first?

Tibo @thsottiaux

2 days ago

Reminder that you can use the Codex App, CLI and SDK with any open source model, not just with OpenAI models. developers.openai.com/codex/config-a…

478 703 7K 2.0M 4K

1 0 3 183 1

View Details

Adrian @agentx_Diff

4 days ago

@sun_baiyang 有老哥互关木

0 0 0 2 0

View Details

Adrian @agentx_Diff

4 days ago

Pretending to be a Singaporean, an American, a Japanese, etc. will be declared a failure. We will eventually embark on the KYC compliance path.

Anthropic @AnthropicAI

7 days ago

The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees. The net effect of

13K 26K 88K 91.1M 24K

0 0 1 19 0

View Details

Adrian @agentx_Diff

a week ago

I increasingly feel that human cognitive bandwidth is an incredibly valuable resource. As Agent and AI capabilities continue to improve rapidly, one person can now run many sessions, tasks, and information streams in parallel. But because of that, we are also forced into more intense context switching, denser decision-making, and more sustained deep thinking. In this environment, what really matters is the ability to keep our minds clear, focused, and efficient. AI can amplify execution, but the clarity of the human mind determines whether those capabilities can be effectively organized and used. So protecting our brains will become increasingly important. Reducing unnecessary mental drain, and leaving enough space for rest and deep thinking.

2 1 4 58 0

View Details

Adrian @agentx_Diff

2 weeks ago

I’m so hard for you, and I need you right now.

Lentils @Lentils80

2 weeks ago

🚨Major Scoop: The first Claude 5 model, Claude Fable 5 (Mythos-class model) is gonna release very soon! It's the same underlying model as Mythos but with increased guardrails, headed to public release

60 127 2K 531K 327

0 0 0 41 0

View Details

Adrian @agentx_Diff

2 weeks ago

Codex side chat is a “side research room” for follow-ups, web/deep research, option comparisons, and fact-checks without interrupting the main working thread. Main thread = construction site. Side chat = research room.

OpenAI Developers @OpenAIDevs

2 weeks ago

Side chats now stay visible when you go full screen in the Codex app. Notifications and unread badges route with the right context, AppShots and voice mode use the active composer, and Escape stops the chat you meant to stop.

13 14 444 31K 44

0 0 0 39 0

View Details

Adrian @agentx_Diff

2 weeks ago

@hellopanghe @niopub I m back

0 0 0 21 0

View Details

Adrian @agentx_Diff

2 weeks ago

Um, just be nice to your Codex account, you know, use it responsibly, alright.

16 0 76 16K 5

View Details

Adrian @agentx_Diff

2 weeks ago

@OpenAIDevs Codex side chat is a “side research room” for follow-ups, web/deep research, option comparisons, and fact-checks without interrupting the main working thread. Main thread = construction site. Side chat = research room.