This isn't exactly a new model but it deserves recognition. A 3B Model called VibeThinker, beats models like Opus 4.5 and Gemini 3 Pro. While these aren't the latest frontier models, it gives us an idea of how open source will catch up and then beat frontier models while also being able to run on consumer level hardware, maybe not now, not in the next year either, but within the next 5 years? Definitely possible. It was built on top of Qwen 2.5 3B (Not even the latest low parameter qwen model either), and focused directly on reasoning tasks. Now the use case this model provides is very specific, it's mainly meant for generating outputs that have a direct Yes/No as an answer. Now that brings up the question, for most of the ICM systems where you aren't directly creating new software or having AI do insane tasks such as editing videos from scratch with animations and such, how far are we from removing frontier models entirely from our ICM systems? I would say that setting up the ICM folder and workflows inside of it can be given to one of the top models, but if the ICM is set up properly, it's a step by step workflow that doesn't change, the results do, but the steps to make the results don't (assuming this under the average use case of the ICM). So if a model is able to achieve something like this, we aren't that far from having tiny hyper focused/fine-tuned models or a cluster of tiny models for a specific use case being implemented into our daily lives. A model for content writing, a model for deep research, a model for front end HTML visual explanation pages. That's exactly what I'm on a mission to build, let's see how it works out. Might be a cluster of fine tuned models, or just a really smart MoE model.

Spare a token, sir? 🪙 Y'all remember two weeks ago when Fable was included in our plans? Just sitting there. Free. Like water from a fountain. Anthropic said nah, that fountain is a vending machine now. $10 per million input. $50 per million output. Double Opus. And every Fable call in Claude Code burns your plan credits at 2x the rate, so that weekly quota you thought was generous? Gone by day 3. I give it about a month before the Clief Notes community looks like this: "Hey man... you got any Fable tokens you're not using? I just need like 200k output tokens. I'll pay you back Friday I swear." "Bro I haven't consumed Fable in three days. I've been surviving on Sonnet 5 like it's ramen." "I used to have a Max plan. A FAMILY. Now I'm out here refreshing the Claude Console watching my usage credits tick down." There's gonna be two types of people in this community. The ones who set a monthly spending cap, and the ones who find out what $50 per million output tokens actually feels like when Fable decides to think for 40,000 tokens before answering your question. That's two dollars just for the thinking. The THINKING. It didn't even answer yet. And don't get me started on the agentic loops. You let Fable run a Claude Code session with dynamic workflows, the thing fans out into a bunch of subagents, and suddenly your invoice looks like you took the whole squad to dinner and everybody ordered the lobster. Meanwhile everyone else is out here watching through the window. They never even got a taste. The real ones already know: The ICM reduces token wasting, Opus 4.8 is included in your plan. Sonnet 5 is $2/$10 intro pricing through August. The local model swarms/subagents run on hardware you already own. Save Fable for the tasks that actually need it, architecture decisions, the hard debugging sessions, the moments where you need the gold standard. Everything else? Run it local and always make sure your ICM is set and engaged. Run it cheap. Hoard those Fable tokens like they're the last clean water in the wasteland.

I want to do the simple thing where someone: - Drops their name and email on my site. - Automatically gets a free resource/lead magnet. - Lands on my email list, tagged and organized, so I can send them stuff later. I've seen people use Supabase for database management (which might be overkill for my current needs) along with Resend for email and lead magnet delivery. What are you using to make that work? I've been using Kit, but frankly, it sucks. I'm wondering what people in this community are doing. I'm not sure if I should switch to another service like beehiiv, for example, or just build something myself.

I hope some members here have some UI design experience and have run into this issue before. So I use Claude Design and Open Design to create layouts for the web apps I build. Both tools generally design the interfaces with React Components. All cool, I use these as prototypes. The problem is I do not use React in my projects so I need to convert these designs to my stack. And this is were I run into a problem. No matter what I do during the conversion the designs are approximated and are not the same design by a long shot. I am stuck. If I can't figure out the prompts to make this work I cannot setup a ICM system to make this work reliably. My experience is literally the reference design is just a suggestion and not a source of truth no matter what prompts I use. This is the same for Codex and Claude Code. How do I fix this process? Anyone has encountered this before?

Figma... I just don't care for it, it's great for collaboration amongst a team, etc, but I don't want to be confined to it's structure and limited capabilities. For every new tool that can work a bit faster, you always lose one primary component: CONTROL.... and Adobe Illustrator, what I can say, they try... but you end up spending more time managing layers than you do drawing. So, the real KING remains throned and always will; Corel DRAW, the absolute total answer to everything vector. This Legacy platform is central to everything I do. Design, planning, everything starts here. Here's a shot of my custom planning board design. I gave you only a piece of the actual workflow view, so you can get the idea, but there's so much I can't show you. A complete schematic of the entire logic before one line is written. A visual roadmap to success for every project. This is the way (if you can draw). (btw, I don't even view layers in my workspace GUI, in Corel Draw it's useless and unnecessary.)