@Michaelzsguo Yes have the same feeling about DeepSeek with Codex harness. 1M tokens is $0.87 though, so how could you spend just 1.85 ?
English
Mark Sergienko
1.2K posts
Mark Sergienko
@Mkrn
Founder https://t.co/mbZcAaOgal https://t.co/l2gKOjUtlf, LLM whisperer.
Lisbon Katılım Ocak 2009
681 Takip Edilen312 Takipçiler
My Deepseek V4 Pro agent (inside codex) has been pursuing goal for more than 13 hours, burning ~100M tokens, and has only costed me $1.85. Yes you saw it right. Not $185, but 1 dollar 85 cents.
English
Using baby names DB to find my new agent name Kudos to github.com/dxdc/babynames
English
@graslogamer @neural_avb @OpenAIDevs @thsottiaux Cool! Share your work !
Also I use this $100 plan in OpenClaw quite extensively and not hitting a limit yet..
Although I'm sure Codex will tighten it up.. I suspect they auto-control how hooked you are and how ripe you're to pay more, based on all the usage signals
English
@Mkrn @neural_avb @OpenAIDevs @thsottiaux I finally started hitting my $200 plan limits. Looking at picking up a second
English
Someone from the Codex team, can you kindly explain the TOS around this?
If I have a $40/mo budget, is it safe to get two Plus accounts?
@OpenAIDevs @thsottiaux
AVB@neural_avb
Told my girlfriend, "current $20 Codex sub is not enough. I need to go $100/mo" She suggested, "why don't you open a second $20 Codex account" I feel so dumb🤔
English
@theo OpenAI Agent SDK for the win! Works amazing with any model, like Deepseek v4 pro !
English
I can't help but feel personally burned by the Claude Code changes announced today.
We put so much work into wrapping the (atrocious) Claude Agent SDK in T3 Code. It was the ONLY path they supported, so we made it work. It was hell.
Now our users are getting their rate limits cut by 40x, despite us doing everything right.
I listened to the Claude Code team. I had my issues with their direction, but I trusted them and took them at their word.
I will never make that mistake again.
Until we see significant change, it is safe to assume any statement from an Anthropic employee is a lie on a timer.
The rug will be pulled, no matter how many promises are made beforehand.
English
@Mkrn @neural_avb @OpenAIDevs @thsottiaux Just remember you’re currently getting double limits on the $100 plan so be prepared for it to drop and the end of May.
English
I'm giving away five $200 Max plans for @DevinAI
If you're using tools like Claude Code, Codex, or Cursor and haven't tried @DevinAI, comment below with what you're building to be eligible. ⚡️
ellis@DriscollEllis
all you nerds hyperventilating about your claude code config would be gobsmacked by the sophistication devin has had for months. cc is catching up but still relatively unserious
English
This is epic! People complaining in comments don't realise in a few months this may very well be profitable.
Chris@chatgpt21
Codex made me money without me doing anything.. Huge turning point for me today, I asked Codex to go off and make me $5. It went out, found a small open-source security/audit bounty path, made a legit PR, followed up with the maintainer, kept my payment details private - (without me asking), handled the GitHub proof/verification loop, and got the work merged. it spent about 22 hours working on multiple security audits. Today I received my first payment from that experiment: $16.88. That’s a $506.40/month run-rate if repeated daily. Not life-changing money yet, but it's deeply exciting to live out Sam Altman's vision for AI, where it will just go out and make money for you. It's awesome to start to see the beginning of that.
English
We finally convinced @eastdakota to let us do billboards in SF. I can’t wait to see them in person later this month.
English
@jamespotter I moved to self-host Plausible with custom events + OpenClaw accessing it and making any kind of report for me
English
I'm paying $180/mo for Mixpanel but I completely stopped using the UI, I just ask Claude questions or get it to render charts and funnels on-the-fly.
They only retain your data for 2 years unless you sign an enterprise deal. It makes less and less sense to use these platforms.
I feel if someone builds a rock solid API-first analytics service with a compatibility layer for the Mixpanel/Amplitude ingest libraries, that would do quite well.
English
DeepSeek v4 pro is very good for OpenClaw but I have a feeling Xiaomi MiMo-V2.5-Pro will be better
English
Testing Cloudflare Codex app, it's critical others catch up (@Railway)
English
We need hooks in Codex = i.e. every frontend change get copy critiqued by Gemini 3.1 pro
English
GPT5.5 Will spill all of your internals right into customer facing frontend copy - prompts, your requests and everything. It doesn't do evaluation on what copy is best for customers unless you separately ask for it
English
@romainhuet Google Ads, Google Search Console, Google Analytics, Plausible analytics, Resend, Intercom
English
We’re thinking about the next wave of Codex plugins.
What’s one you’re missing today?
English
@enjojoyy Will you be paying for replit? Will it give you dedicated IP and firewall controls if your agent decide to build something on it's box? Will it have enough RAM and CPU?
Also looking for an easy setup for a presentation I'm giving to non-techies soon
English
Update: it took me 2 minutes to spin off Hermes with Replit.
I honestly forgot Replit exists but so far it’s the fastest and easiest option for someone who never used SSH.
albina@enjojoyy
What's the easiest way to host an autonomous agent for a non-technical person currently? (Openclaw, Hermes, Pi etc.)
English
@thsottiaux There needs to be a version control alternative to git that remembers every LLM's change and can roll them back individually, train on these individual changes and undos to make the model better!
English
