See 120B LLM Running Locally on an Old PC in One Continuous Take No cloud. No GPU upgrade. No expensive computer. 😎 We just ran OpenAI GPT-OSS 120B on a computer from 14 years ago — powered by Tiiny AI Pocket Lab. See what personal AI actually looks like.

Join us at CES 2026 Time to try Tiiny AI Pocket Lab in person! We're setting up 4 hands-on demo zones: ⚡ 1 running GPT-OSS 120B fully offline 🕹️ 3 local agent experience zones Come experiment, and see how it runs in the real world. Love to meet you there.👋 #AI #LLM #OpenAI

😀Here's the link:

Wonder how Tiiny runs a 120B LLM locally? The key is PowerInfer — a heterogeneous inference architecture that assigns "hot neurons" and "cold neurons" across NPU/CPU for unprecedented 120B decoding speed on consumer hardware. Code & details on GitHub👇 #AI #LLM #OpenAI

Meet Tiiny — the World's Smallest Personal AI Supercomputer Certified by Guiness World Records. 😎 Run GPT-OSS 120B model offline with 20 tokens/s ⚡️ Empowered by PowerInfer & TurboSparse, delivering full intelligence ✨ No cloud. No token fees. AI in your hands. #AI #LLM

What if your AI‑assisted tasks go deeper?⚙️ If your workflow gets a bit more complex, the story with open-source LLMs doesn't really change. Most real‑world tasks still fit comfortably within these model size ranges. Here's the second snapshot. #TiinyAI #AI #LLM #OnDeviceAI

Not sure what open-source LLM model size you need?🤓 Tried bringing AI into your workflow? Reasoning, speed, consistency, and cost all factor into choosing the right model size. It's all about matching the model to the job. Here's a quick snapshot to help. #TiinyAI #AI #LLMs

Not another AI startup. We’re building something real — private, local, and yours.

Building something small, smart, and completely offline. Private AI shouldn’t live in the cloud.

🎅 😂