Samuel Fajreldines @devindolar

@JulianGoldieSEO Putting AI agents in every channel makes permissions the first test. Each agent should see only the messages it needs, and publishing should still require review.

0 0 0 9 0

View Details

Tibo @thsottiaux

3 days ago

Back at the laptop. The usage limits have been reset for all paid users of Codex and ChatGPT Work. Weeeeeeeee. It's a good day!

We’re celebrating the fast adoption of chatGPT Work and all the incredible effort that went into it today. I’m feeling like a limit reset. Hold on tight to your ultra and /fast and see you in a few hours when I’m back at the laptop!

1K 401 11K 1.7M 511

2K 599 13K 1.2M 336

View Details

Samuel Fajreldines @devindolar

3 days ago

@FurkanGozukara remotecode.io trims repeated context :)

0 0 1 22 1

View Details

Samuel Fajreldines @devindolar

3 days ago

@saeed_vz Use remotecode.io and save usage :)

0 0 0 8 0

View Details

Samuel Fajreldines @devindolar

3 days ago

@iAmHenryMascot Burning through five Codex weekly limits in under a week is enough reason to run repetitive agent loops locally and save Codex for the hard decisions.

1 0 0 14 0

View Details

Samuel Fajreldines @devindolar

3 days ago

@torosgo @OpenAI @thsottiaux remotecode.io and save tokens :)

1 0 1 1K 5

View Details

Samuel Fajreldines @devindolar

3 days ago

@piyush784066 cry

0 0 0 6 0

View Details

Samuel Fajreldines @devindolar

3 days ago

I can confirm. Codex as orchestrator and Grok as executor is the best stack right now!

NIK @ns123abc

4 days ago

Grok 4.5 is literally the best value per dollar + super heavy subscription is 2x cheaper ($100) with nearly DOUBLE the rate limits than both codex and claude 20x max plans

37 25 367 26K 39

0 0 0 39 0

View Details

Samuel Fajreldines @devindolar

3 days ago

@omkarships 103B in codex. Still have claude and grok And sometimes I use deepseek.

0 0 0 49 0

View Details

Samuel Fajreldines @devindolar

3 days ago

@anichebedennis Codex

1 0 1 187 0

View Details

Samuel Fajreldines @devindolar

3 days ago

@rajgupta7813589 For unlimited usage I'd pay.

0 0 0 25 0

View Details

Samuel Fajreldines @devindolar

3 days ago

@betomoedano remotecode.io and use insomnia :)

0 0 0 3 0

View Details

Paweł Huryn @PawelHuryn

4 days ago

I ran 7 frontier models against 105 hidden bugs in two unrelated codebases. Strict fixes. Either fixed or not. GPT-5.6 Sol 31 Fable 5 24 Opus 5 21 Kimi K3 21 Grok 4.5 16 Opus 4.8 9 Sonnet 5 9 Fable 5 lost to Opus 5 on the first repo, 9 to 11. Won the second, 15 to 10. GPT-5.6 Sol consistently crushed everything.