Introducing `gpt-oss Pro Mode`! Basically o3 Pro, but for the new OpenAI open source models :) Pro Mode chains together up to 10 instances of the new OpenAI GPT-OSS model, enabling it to produce a better answer than one instance alone could create! Open source, link here:

On Macbook M4 Max gpt-oss 120b: 48 tok/s high 35 tok/s normal 23 tok/s low power Very nice outputs, new model of choice

A voice agent powered by gpt-oss. Running locally on my macBook. Demo recorded in a Waymo with WiFi turned off. I'm still on my space game voice AI kick, obviously. Code link below. For conversational voice AI, you want to set the gpt-oss reasoning behavior to "low". (The default is "medium".) Notes on how to do that and a jinja template you can use are in the repo. The LLM in the demo video is the big, 120B version of gpt-oss. You can use the smaller, 20B model for this, of course. But OpenAI really did a cool thing here designing the 120B model to run in "just" 80GB of VRAM. And the llama.cpp mlx inference is fast: ~250ms TTFT. Running a big model on-device feels like a time warp into the future of AI.

I built a real estate AI Agent team with multi-agents using gpt-OSS and Firecrawl. It has 3 AI agents working together as a team: 1. Property Search Agent 2. Market Analysis Agent 3. Property Valuation Agent 100% Opensource with step-by-step tutorial.

openai gpt-oss chat app with thinking-ui (100% local):

Je to jen něco málo přes 19 hodin, co OpenAI konečně upustila od gpt-oss, plně open source modelu. A lidé už jsou šíleně kreativní a mají s tím výsledky. 10 divokých příkladů: