Adel

AI in general

[Adel @IrregularAdel]

@kimmonismus cool! congrats!

[Adel @IrregularAdel]

@KinetixSEO https://t.co/94e16JITNi

[Adel @IrregularAdel]

Can’t believe we got accepted, 5x pro plans and $1000 on credits for our small team - thanks Anthropic! gonna cook https://t.co/94e16JITNi

[Claude @claudeai]

The Claude Startups program is expanding to more founders.

Members can get a year of Claude Team, API credits, special offers on the tools a startup runs on, and office hours with Anthropic’s Applied AI team.

[Adel @IrregularAdel]

@thsottiaux So, reset tomorrow?

[Adel @IrregularAdel]

@kimmonismus @thsottiaux https://t.co/ImrVrFJG5W

Can we use this

[Adel @IrregularAdel]

@kimmonismus Also, I love Codex app, but somehow it became bloated and overloaded just after DevDay

[Adel @IrregularAdel]

@pradeepXkapoor Imo they slowed down everything in favor of fast modes. That’s torture

[Adel @IrregularAdel]

Codex became EXTREMLY SLOW for me, that’s so frustrating. I use Sol and Luna mostly, both are just slowpokes. Is OpenAI rerouting compute ?

[Adel @IrregularAdel]

@sama Plus users?

[Adel @IrregularAdel]

@AestherML They are largest startup ever, of course they plan as they go - look at sora app, atlas, and many other features and services they pivoted from

[Adel @IrregularAdel]

@kimmonismus To be fair, he didn’t say $20 plan is halved

[Adel @IrregularAdel]

@OpenAIDevs From the Netherlands! Building https://t.co/94e16JITNi

[Adel @IrregularAdel]

@kimmonismus How would you compare it to Astra?

[Adel @IrregularAdel]

@kimmonismus Plus plan got robbed?

[Adel @IrregularAdel]

@araseb_ make money - https://t.co/94e16JJrCQ

[Adel @IrregularAdel]

Did Opus 5.5 just cut OpenAI releases in the corner? And no sol/luna/astra-minor today AND no resets?

[Adel @IrregularAdel]

@kimmonismus @bridgemindai Could still be larger than Sol, but not as hungry as Astra

[Adel @IrregularAdel]

@andrewgwils What if conferences used review credits? Reviewing others earns credits; submitting consumes them. Everyone gets a basic free allowance, but heavy submitters must contribute more. Opting into AI review costs fewer credits. Access stays open; human review is scarce.

[Adel @IrregularAdel]

@ziv_ravid @chrmanning How would anyone be able to reproduce gazzilion hours GPU training run?

[Anshu @anshuc]

I had Astra analyze all my Codex threads and learned a lot about quota. Here's how to get the most Astra usage. (applicable to Fable too)

I've spent ~$5000 on Astra since release*:
- 70% is JUST cached input reads
- 16% is uncached input
- 13% is output
- <1% is cache writes

Till now, my Codex threads kept getting longer and longer, but Astra broke the trend (see chart).

Long context is eating most of your quota. The biggest levers:
- Smaller threads. New thread for each little task. Don't let context build up
- Use smaller models that delegate coding tasks to Astra subagents. Don't let Astra see big outer thread
- Avoid continuous polling. Have Astra set timers and wake back up for long-running tasks
- Trim tokens wherever you can: skills/MCPs, AGENTS.md, custom instructions, etc
- 3rd party tools like RTK that return more compact outputs might help (haven't tested this yet)

Investigating that 70% input cost in more detail:
- 40% is encrypted content (reasoning or other OpenAI stuff)
- 20% is file reads
- 10% is tool call arguments, schemas, etc
- 7% is skills (catalog, loading instructions, output)
- 5% is base instructions (system prompt, custom instructions, AGENTS.md)
- 5% is browser/computer use input
- 3% is command/tool output
- The rest is minor or not actionable: compaction handoffs, assistant prose, user messages

What you can do:
- Use the lowest reasoning you can for your task, to cut CoT tokens
- Make your repo as agent-friendly as possible. Small files/modules with clear folder structure, so Astra isn't grepping and reading large chunks. Try asking Astra to delegate file search tasks to subagents like Luna
- Don't dump in tons of skills, instructions, AGENTS.md stuff; be intentional about it
- Delegate browser/computer use to smaller models unless you need max performance

And, probably goes without saying:
- Don't use /goal with Astra. The few times I did absolutely obliterated my quota
- /fast mode basically only if Tibo says he's resetting in an hour and you have 100% quota to burn

Let me know if you have any observations to share!

*I didn't literally drop $5k, this is mapping token to equivalent API prices