I build agentic kits that help people and businesses work faster, get more clients, and stay ahead. Productivity + acquisition systems, no code.

Joined January 2021
Include

Only show posts containing:

Exclude

Hide posts containing:

Time range
-
Minimum likes
Replying to @LuminaBench
😂😂 their distilled model mogs the teacher crazy
2
136
It needs warmup and understands your app once done , it will cook like crazy
1
2
180
Did you use even one neuron before writing this post, or are you just a moron?
2
271
Replying to @Da7_Tech
Let me just block you, because all you do is yap without bringing a single number to back up your claims. Astra hasn’t even been released yet.
1
1
4
526
So opencode go 10$ = Ollama 20$ someone to explain this
well, Ollama just changed the rules. they removed the 5-hour resets and weekly caps entirely new pricing: Pro: $20/mo → $60 usage included Max: $100/mo → $300 usage included Team: $500/mo → $1,000 shared usage so now you can use what's included, then pay per token at the same rates if you go over no more "hit the wall, wait until next week." you know how much you have, what you used, and what happens after one important detail: old subscribers stay on the old plans unless they switch, which means they keep the old session + weekly limits honestly, it’s very transparent and usage looks pretty generous. whether it works out cheaper depends on how much you use after that curious how long before anthropic and openai feel pressure to do the same.
1
67
I packaged my TikTok slideshow factory. Not a prompt. An installable Hermes system: → 3 focused agents → visual identity lock → Cloudinary media memory → publishing wired in → 30-day adaptive workflow No face. No filming. No rebuilding the stack from scratch. €87 one-time (was €129) kitstore.macorha.com/en/p/ti…
2
22
Replying to @nahcrof
I know this can be very personal, but it would really help us if you could share a little more about the issues you faced. We’re here to support you, and of course, we completely understand if you don’t feel comfortable sharing more personal details. Either way, thank you again for everything you’ve already shared and done.
1
10
1,812
I really want to see one day a computer made by OpenAI and deeply integrated with AI, much more than what Microsoft did
LMAO the Codex reset lore somehow gets even better Tibo says the resets aren’t coordinated with marketing or finance He can literally reset everyone’s usage whenever he thinks it feels right AND THERE IS NOW AN ACTUAL PHYSICAL RESET BUTTON 😭 OpenAI may have accidentally invented the most powerful button in AI
2
48
Replying to @nahcrof
I hope I built my Saas on you
10
878
Replying to @nahcrof
Bro, with pricing like this, are you trying to cause an outage right now? 😭 People are going to rush in, overload the service, and the quality will take a hit
1
12
3,448
Replying to @jenzhuscott
It’s just too good
2
2,167
ChatGPT has been such a great experience. I can't believe we only paid 20 bucks for it! My aunt struggles a bit with writing, so I just asked Sol to create a typing practice website for her. It focuses solely on helping her improve her typing speed and grammar, and that's it. @thsottiaux @sama kudos yall
3
16
opencode is unsusable ! Deepseek really cooked
3
185
@grok explain this image
1
1
825
Replying to @rezoundous
More than that, they do not have any better model in the bag actually, that say a lot
4
1,678
Grok 4.5 sits in that painful middle: not cheap or smart enough to be the obvious choice, not premium or brilliant enough to justify the cost.
2
2
14
1,268
It isn’t the correct list : Opus: for heavy work use GLM 5.2 and DeepSeek v4 pro max daily Sonnet: DeepSeek v4 flash max Haiku: Ornith 35B Try this
My current theory is most people (that aren’t say doing bleeding edge research etc) could get away with a setup like: Opus: GLM 5.2 Sonnet: MiniMax M3 Haiku: Qwen 35B Which is great on paper, but also needs just shy of 3TB to deploy ish.
2
121