AI & Machine Learning Threads

@PrismML

Thread

Today, we’re announcing Bonsai 27B: the first 27B-class model to run on a phone. Bonsai 27B is the new multimodal flagship of the Bonsai family. Based on Qwen3.6 27B, it brings a new capability tier to local AI: multi-step reasoning, structured tool use, long-context workflows, and coherent agentic...

@RitOnchain

Article

Every quant strategy has a dirty secret: it only works in one regime. A momentum strategy crushes in trending markets and bleeds in choppy ones. A mean-reversion strategy prints in sideways markets and gets destroyed in trending ones. Most quants discover this the hard way - live, with real capital....

@CliffordSosin

Article

Superintelligence arrived. You probably didn't notice, because it turned out to be kind of incremental....

@deredleritt3r

Thread

Things we know about OpenAI's unreleased internal model: - When given enough compute, the model can solve the unit distance problem 48% of the time, fully autonomously, in one shot, without using Lean, without using any special harness. (How much compute? OpenAI doesn't say. But I doubt that OpenA...

@DeRonin_

Thread

4 HOURS COURSE ON HOW TO RUN 1-PERSON AI BUSINESS JUST DROPPED worth every minute if you have the time if you don't, I wrote a full course on the same topic from running my own $40k MRR solo agency: every system, prompt and price 20 minutes to read, free, below ↓ <span class="quote-card" data-qu...

@andreysuperior

Thread

The companies charging $200/month for AI assumed local hardware would never catch up. It caught up. He's assembling the proof right now. A PC that runs Qwen 3.6 27B a free model that beats Claude on vision by 7 points. DeepSeek R1 for math and reasoning. Llama 3.3 70B for everything else. All loc...

@h100envy

Article

Not about speed. About making several agents with different viewpoints argue over one decision and reach a conclusion better than any of them alone. With full code for the orchestrator, experts, and merge....