09:02 Tutorials21 hours ago Needle: Finetune a 26M Tool-Calling Model Locally with Ollama Needle is a 26-million-parameter encoder-decoder transformer with one job: given a natural language request and a list of available t... 0 comments 1.9K views
09:01 Foundation Models2 days ago NVIDIA’s Two-Tower Model Generates Text 2.4x Faster Without Losing Quality NVIDIA has released NeMo-Tron 2 Tower 30B A3 Base, a 30-billion parameter model that rethinks how large language models generate text... 0 comments 1.1K views
10:50 Research & Benchmarks2 days ago Laguna XS 2.1: Poolside’s Local Coding Agent Tested – Nine Languages Fahd Mirza puts Poolside's newly released Laguna XS 2.1 through a live evaluation using the Hermit agentic framework. The model is a... 0 comments 1.2K views
12:40 Research & Benchmarks3 days ago Sonnet 5 vs Ornith 35B: Can a Local Model Beat Closed-Source? Fahd Mirza runs a direct head-to-head evaluation between Claude Sonnet 5 and Ornith 1 35B — a new open-source mixture-of-experts codi... 0 comments 4.3K views
13:29 Coding & Dev Tools4 days ago Control What Your AI Agents Can Do: Archestra + Ollama Hands-On Fahd Mirza walks through Orchestra (stylized as Archestra), an open-source platform built for running AI agents safely in production,... 0 comments 1.3K views
11:00 Coding & Dev Tools5 days ago Ornith 1.0 35B in GGUF – Beats Models 10x Its Size – Run Locally Fahd Mirza puts Ornith 1.0 35B through its paces in this hands-on local deployment walkthrough. Ornith is a mixture-of-experts model... 0 comments 6K views
13:28 Tutorials6 days ago Krea 2 Turbo OSS Text-to-Image Model: Run and Test with ComfyUI Fahd Mirza installs and systematically tests the Krea 2 Turbo, a 12-billion-parameter open-weights diffusion transformer from Krea AI... 0 comments 722 views
08:09 Business & Strategy6 days ago Someone Fine-Tuned a Model on 10 Examples in 3 Minutes: Qwable-5 27B Coder Fahd Mirza uses a deliberately provocative model release — Qwable-5 27B Coder — as the centerpiece of a broader argument about credib... 0 comments 2.6K views
14:08 Benchmarks7 days ago Qwythos 9B: When You Train a Small Model on Claude Traces: Run Locally Fahd Mirza introduces and benchmarks Qwythos 9B, a reasoning-focused open-source model fine-tuned on over 500 million tokens of Claud... 0 comments 2.9K views
08:49 Tutorials1 week ago Hermes Agent /learn — Teach Your AI Agent Anything Fahd Mirza demonstrates the new `/learn` command in Hermes Agent, the open-source AI operator from Nous Research that runs in a termi... 0 comments 1.4K views