Models & Labs
NVIDIA Unveils RTX Spark Laptops and DGX Stations
Lev Selectorhigh confidence
Why it matters
- →Provides powerful local computing options for AI development.
- →Supports the execution of large models without cloud dependency.
- →Enhances accessibility and performance for AI researchers.
NVIDIA has launched its latest hardware offerings, the RTX Spark laptops and DGX Stations, designed to support the local execution of large AI models. These devices aim to provide enhanced computational power and efficiency for AI developers and researchers, facilitating more robust and scalable AI solutions.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
How we got here
- 1NVIDIA Nemotron 3 Available on Together AI
Together AI Blog · March 11, 2026 · Background
- 2Nvidia GTC Reveals New AI Chips
Lev Selector · March 20, 2026 · Related
- 3Together AI Launches NVIDIA Nemotron 3 Nano Omni
Together AI Blog · April 28, 2026 · Background
- 4NVIDIA Launches Nemotron 3 Nano Omni
Matt Wolfe · May 1, 2026 · Related
- 5NVIDIA Unveils Nemotron 3 Ultra Model
Sam Witteveen · June 4, 2026 · Background
- 6NVIDIA Unveils Nemotron 3 Ultra for AI Agents
Matt Wolfe · June 5, 2026 · Related
What happened next
- 7NVIDIA Unveils RTX Spark for AI-Driven Gaming
NVIDIA Blog · June 7, 2026 · Same story
- 8NVIDIA Unveils Nemotron Lightning Model
Sam Witteveen · August 11, 2026 · Background
- 9NVIDIA Unveils Nemotron 3.5 Lightning
Lev Selector · August 14, 2026 · Related
- 10NVIDIA Unveils RTX Spark for Next-Gen Gaming
NVIDIA Blog · August 25, 2026 · Same story
- 11Nvidia Unveils AI-Powered RTX Spark PCs
WIRED AI · September 3, 2026 · Same story
Follow this story
Open the full story →NVIDIA Unveils RTX Spark for Local AI Agents
8 developments
- Jun 1 · NVIDIA Blog
NVIDIA Unveils RTX Spark for Local AI Agents - Jun 1 · TechCrunch AI
Nvidia Unveils RTX Spark CPU for AI PCs↳ Nvidia expands RTX Spark to include a CPU for AI PCs with support from Dell, HP, and Microsoft. - Jun 2 · The Rundown AI
Nvidia Unveils AI Agent-Centric Tech at COMPUTEX↳ Nvidia unveils Vera processor, Cosmos 3 robotics model, and Nemotron 3 Ultra at COMPUTEX 2026. - Jun 2 · NVIDIA Blog
NVIDIA and Microsoft Partner on AI Deployment Stack↳ NVIDIA partners with Microsoft to integrate RTX Spark and DGX Station into Windows and Azure, boosting SQL execution speeds. - Jun 3 · WIRED AI
Nvidia Unveils RTX Spark Laptops at Computex - Jun 3 · NVIDIA Blog
NVIDIA Unveils New AI Agent Skills at CVPR↳ NVIDIA unveils new AI agent skills for autonomous vehicles and robotics at CVPR, integrating with Cosmos 3 to streamline research workflows. - Jun 4 · The AI Daily Brief
NVIDIA Launches RTX Spark and Vera Rubin Platform↳ NVIDIA launches the Vera Rubin platform alongside RTX Spark to enhance AI inference and agent-based workloads. - Jun 5 · Lev Selector
NVIDIA Unveils RTX Spark Laptops and DGX Stations (This article)
More in Models & Labs
Models & Labsother
llama.cpp b12340 adds CUDA 13 and ROCm 10.0
This release quietly closes the hardware gap for local inference by adding default support for CUDA 13 and ROCm 10.0 across Linux and Windows. Users with newer NVIDIA GPUs or AMD accelerators no longer need to compile from source to get performance, while Apple Silicon builds now ship KleidiAI disabled by default, shifting the focus back to standard ARM optimizations. The inclusion of Snapdragon support for Linux and Android marks a significant step toward on-device AI beyond the x86/ARM desktop ecosystem. It is less about new features and more about ensuring llama.cpp remains the universal runtime for every available accelerator.
llama.cpp Releases
Models & Labscoding
llama.cpp b12342 adds few-row MMQ for speculative decoding
Speculative decoding just got a significant speed boost on NVIDIA GPUs. This release introduces a specialized kernel for 'few-row' matrix multiplications, targeting the small batch sizes typical of verification passes in speculative decoding workflows. By optimizing these specific operations for quantized formats like Q8_0 and IQ4 variants, llama.cpp reduces latency where it matters most during token generation. The change is CUDA-only for now but leverages hardware capabilities from Ampere onwards, making local inference faster without requiring new model architectures or complex configuration changes.
llama.cpp Releases
© TechCrunch AIModels & Labsmodels
Google deploys Gemini 4 Hydrogen for faster Search
Google is quietly shifting its search infrastructure by replacing older fast models with Gemini 4 Hydrogen, a tier designed specifically for speed and cost-efficiency rather than raw reasoning power. This model completes tasks 2.5 times faster than its predecessor while beating the more powerful Argon on specific intelligence benchmarks, proving that specialized efficiency can outperform brute force in high-volume workflows. By integrating this into Search's fan-out technique, Google now breaks down complex queries into subqueries with greater precision and speed, delivering richer visual responses without the latency of heavier models. This marks a strategic pivot toward agentic execution layers within consumer search, making AI Mode feel snappier for everyday multi-part questions.
TechCrunch AI