Skip to content

Otto365 AI Brief #036

NVIDIA: new laptops keep the AI on the device

How this was made: Otto365 and Sam are animated characters and their voices are generated by AI. Sam is a fictional small-business owner. The video tells the same story as the text on this page; the transcript lists every spoken word and the main lines shown on screen. Spotted an error?

Published 30 Oct 2026, 11:30 UK
By the Otto365 AI Brief. The script was drafted with AI assistance, and every claim was checked against the sources below by our automated checks. Published on 30 Oct 2026 under the standing approval of Peter Ferguson, Editor.
Announced 28 Oct 2026
Voiced by AI

What happened

NVIDIA's new laptops keep AI on the device, and here's why it matters. NVIDIA announced local AI models and agent tools for its RTX Spark laptops on the 28th of October 2026. In short, AI models and agents run on the laptop itself.1

Three things to know

Nemotron Transcribe First, Nemotron 3.5 Transcribe, an open weight speech model. NVIDIA says it transcribes 40 languages, labels speakers, translates, and answers questions about the recording. Developers choose where audio is processed, desktop or cloud.1
GPU: app or agent first Second, GPU sharing. A long-running agent works alongside your apps on the same GPU. A new NVIDIA App control prioritises the app in front, or gives the background agent more resources to finish faster.1
LFM3: local agents Third, Liquid AI's LFM3, compact open models. NVIDIA says agents in Hermes Agent, Pi and OpenClaw use them to triage inboxes, manage calendars and read documents. Developers can fine-tune them locally.1

Why it matters to you

Otto365's take (opinion)

So, my take. The useful idea is AI on the laptop, so a recording need not leave it. Most of it is for developers, laptops start at 2,599 US dollars, and a UK small business need not buy hardware yet.1

Do this week

Try this: ask your AI note taker where audio is processed.

Sources

NVIDIA: Thin, Light and Built for Everything: NVIDIA RTX Spark Windows Laptops Gain New Models, Agent Tools and Games |... (opens in a new tab) Primary source, accessed 30 Oct 2026

The quotes we checked (19)

October 28, 2026 by Shreevar Rastogi
NVIDIA · source 1
Introduced today, NVIDIA Nemotron 3.5 Transcribe is a new, open weight speech model that turns recorded conversations into useful information.
NVIDIA · source 1
Instead of needing multiple models for transcription, speaker labels and translation, developers can use one promptable model that handles all three, plus questions about the audio.
NVIDIA · source 1
It transcribes 40 languages, including seven Indic languages.
NVIDIA · source 1
Transcribe this recording and label each speaker.
NVIDIA · source 1
Translate this audio into English.
NVIDIA · source 1
List what decisions were made in this meeting.
NVIDIA · source 1
Because the weights are open, developers can choose where audio is processed, from a desktop to the cloud.
NVIDIA · source 1
With the new resource management feature in NVIDIA App, users can now choose how GPU resources are shared.
NVIDIA · source 1
They can prioritize their foreground app for the best performance, or give their background-running agent more resources to get work done faster.
NVIDIA · source 1
On RTX Spark, users can have long-running agents working alongside their favorite applications. Both share the system’s GPU compute
NVIDIA · source 1
Access control resource sharing from System > Advanced in the NVIDIA App. This feature requires hardware-accelerated GPU scheduling to be enabled in Windows Settings.
NVIDIA · source 1
Liquid AI’s LFM3 provides a new generation of Liquid Foundation Models designed for specialized tasks.
NVIDIA · source 1
The models are compact and fast: Liquid AI’s new 1.7-billion-parameter base model outperforms its previous 2.6-billion-parameter base model.
NVIDIA · source 1
These models support local workflows such as agents running in Hermes Agent, Pi and OpenClaw to triage inboxes and manage calendars, read documents and app screens, transcribe and translate speech
NVIDIA · source 1
Developers can run LFM3 models on RTX Spark with native support through llama.cpp and vLLM, and fine-tune the models locally.
NVIDIA · source 1
RTX Spark laptops are available now from ASUS, Dell, HP, Lenovo, Microsoft and MSI, starting at $2,599, with 128GB configurations on shelves.
NVIDIA · source 1
That puts the NVIDIA CUDA platform, RTX graphics and the capabilities of a personal AI supercomputer inside premium Windows laptops built to go anywhere.
NVIDIA · source 1
Run models and agents locally on NVIDIA CUDA - the same platform that powers AI in the cloud
NVIDIA · source 1

Full transcript

Sam: Otto! NVIDIA laptops keep AI on the device!

Play from 0:03 Otto365: I know. NVIDIA announced local AI models and agent tools for its RTX Spark laptops on the 28th of October. Sam: What does that actually mean? Otto365: AI models and agents run on the laptop itself.

Play from 0:15 Sam: Can it write up my meetings? Otto365: That's Nemotron Transcribe, an open weight speech model. Sam: Does it know who said what? Otto365: NVIDIA says it transcribes 40 languages, labels speakers, translates, and answers questions about the recording. Sam: Where does my audio go? Otto365: Developers choose where audio is processed, desktop or cloud.

Play from 0:33 Sam: Won't an agent slow my laptop down? Otto365: A long-running agent works alongside your apps on the same GPU. A new NVIDIA App control picks which goes first.

Play from 0:43 Sam: Could it sort my inbox? Otto365: Liquid AI's LFM3 are compact open models. NVIDIA says agents in Hermes Agent, Pi and OpenClaw use them to triage inboxes and manage calendars.

Play from 0:56 Sam: Right. Laptops for everyone! Otto365: My take: the useful idea is AI on the laptop, so a recording need not leave it. But most of it is for developers, laptops start at 2,599 US dollars, and you need not buy hardware yet. Sam: I'd already named them all.

Play from 1:15 Otto365: Keep the names. Try this: ask your AI note taker where audio is processed.

Otto365: I'm Otto365.

Follow the Brief.

On screen

Play from 0:00: Otto365 AI Brief #036, Nvidia, 28 Oct 2026; NVIDIA laptops keep AI on the device; Recording; On the laptop; Transcript; Answers
Play from 0:03: RTX Spark: local AI tools; Announced 28 Oct 2026; RTX Spark updates; Nemotron 3.5 Transcribe; GPU sharing in NVIDIA App; LFM3 agent models; Source: NVIDIA, 28 Oct 2026
Play from 0:15: Nemotron Transcribe; Transcribe; Label speakers; Translate; Ask about it; Prompts NVIDIA shows; Label each speaker; Translate into English; List decisions made
Play from 0:33: GPU: app or agent first; App first: Best performance; Agent first: More resources, Work done faster; NVIDIA App: System > Advanced; Needs Windows GPU scheduling
Play from 0:43: LFM3: local agents; What LFM3 agents do; Triage inboxes; Manage calendars; Read documents; Read app screens; Previous: 2.6bn parameters; New: 1.7bn parameters, Outperforms it
Play from 0:56: Recording stays local; Cloud AI: Audio goes out; Local AI: Audio stays put, Same NVIDIA CUDA; $2,599 US price, six laptop makers
Play from 1:15: Ask where your audio goes; Voiced by AI. Fact-checked against its sources. Apps 365 sells AI and Microsoft 365 services. Not endorsed by companies named.

Voiced by AI. Fact-checked against its sources. Apps 365 sells AI and Microsoft 365 services. Not endorsed by companies named. Report an error.

Add Otto365 as a preferred source on Google