NVIDIA published a technical blog exploring how to run open reasoning models locally on NVIDIA Jetson, including model selection, NVFP4 quantization, speculative decoding, and practical validation for real workloads.
Public source
Publisher name
Public post
Frontier-class reasoning is moving to the edge. ⚡ A year ago, running advanced reasoning models at the edge felt out of reach. Today, the question is no longer relevant…
Company
NVIDIA
- Industry
- Computer Hardware Manufacturing
- Location
- Santa Clara, US
- Company size
- 10,001+ employees
About NVIDIA
Since its founding in 1993, NVIDIA (NASDAQ: NVDA) has been a pioneer in accelerated computing. The company’s invention of the GPU in 1999 sparked the growth of the PC gaming market, redefined computer graphics, ignited the era of modern AI and is fueling the creation of the metaverse. NVIDIA is now a full-stack computing company with data-center-scale offerings that are reshaping industry.
See moreLatest activity
Latest activity from NVIDIA
516 signals
Presence & Recognition
NVIDIA is hosting the GTC Berlin conference in Berlin from October 20 to 22, 2026.
Products & Services
NVIDIA released NCCL Extensions 0.1.0, a home for hot and emerging AI communication patterns built on top of NCCL.
Products & Services
NVIDIA introduced RoboTok, a new technology that uses actor-centered 3D hand trajectories to retrieve human demonstrations across variations in viewpoint, scene appearance, and occlusion for robot learning.
Discover more
Similar signals
Similar public activity from other companies.
Research & Knowledge
Tensordyne
Tensordyne published a whitepaper on how it solved for what really matters in AI inference.
Research & Knowledge
Intel
Intel and Infosys conducted a joint evaluation of open-source LLM inference on Intel Xeon 6 processors with Intel AMX for real-world AIOps workloads.
Research & Knowledge
RetroGradient Labs
RetroGradient Labs published information about the Jetson AGX Thor edge device with 2,070 FP4 TFLOPS.
Research & Knowledge
SambaNova