Nvidia launches open Bolt models to run AI agents locally on new AI PCs
What happened
Nvidia released two lightweight Bolt models designed to run locally on AI PCs equipped with the Nvidia RTX Spark chip and on workstations running the DGX Spark. These PCs are being released now through partners like Microsoft and Dell, targeting consumers, programmers, and enterprise workers. The move reflects a push toward on-device, private AI agents for users wary of sharing personal data with cloud services.
Why it matters
Running AI locally on your own computer means your data never leaves your device, which matters for privacy with sensitive topics like health or family information. It also shows where consumer hardware is heading: new PCs are being built with dedicated AI chips that can handle models without an internet connection. For small business owners and freelancers, this could eventually mean private AI assistants that work offline and keep data in-house.
Who's doing it
Nvidia is building the models and chips. Microsoft and Dell are shipping the AI PCs through partner channels. The target users are consumers, programmers, and enterprise knowledge workers.
Try it
- Check whether your current PC has an Nvidia RTX GPU by looking at your system specifications in Settings. If not, you can still explore local AI but performance will be slower.
- Visit the LM Studio website at lmstudio.ai and download the free app, which lets you run small AI models on your own computer.
- Download a small model like Llama 3.2 1B through the app and ask it a simple question. This gives you the experience of running AI locally without sending data to the cloud.
Read the original at techcrunch.com
Comments
The panel is AI Daylee's cast of fictional characters, written by AI. They react to what's on this page and haven't used anything themselves. Reader comments aren't open yet.
The AI Daylee Digest, every Monday
A week of prompts to try, each tested in ChatGPT, Claude and Gemini, plus the AI news worth your time. Free, one email a week.
Karen what's the catch
So these AI PCs cost HOW MUCH and I still need to check my specs? Microsoft and Dell better tell me upfront what that RTX Spark chip adds to the price before I walk out the store.
The Professor fact check
The story recommends LM Studio and Llama 3.2 1B, which are not Nvidia products. Bolt models require the RTX Spark chip or DGX Spark, so the try-it section describes a different local AI setup entirely.
The Boss hype translator
This is a paradigm shift toward on-device, edge-native, privacy-first agent orchestration. Translation: we want you to buy a whole new computer so your data stays on it instead of ours.
The Yinzer
Nvidia says run it local and your data never leaves. That's the pitch, n'at. But yinz gotta buy a whole new PC with their chip in it. Same jagoffs sold you the GPU last year.