NVIDIA Unveils Smaller Jetson Thor Modules

// 1h agoPRODUCT LAUNCH

NVIDIA Unveils Smaller Jetson Thor Modules

NVIDIA has expanded its Jetson Thor robotics platform with the new Blackwell-based T3000 and T2000 modules, bringing high-performance edge AI to mainstream robotics. The compact modules deliver up to 865 FP4 teraflops of compute and 32GB of memory at half the size and power of the flagship T5000.

// ANALYSIS

Hot take: NVIDIA's launch of smaller Jetson Thor modules indicates that the robotics race is transitioning from a raw compute specs war to a practical optimization challenge of memory and power.

–Power and thermal envelopes of mobile and humanoid robots make full-sized GPU platforms impractical for mass-market deployment.
–The 32GB memory configuration is strategically sized to run optimized, quantized multimodal models on the edge without excessive power drain.
–Offering tiered hardware options enables robot manufacturers to scale compute based on specific application requirements rather than paying for excess capacity.

// TAGS

nvidiajetson-thorroboticsedge-aihardwareblackwellt3000t2000

DISCOVERED

1h ago

2026-07-15

PUBLISHED

1h ago

2026-07-15

RELEVANCE

8/ 10

AUTHOR

dball1126

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

VIDEO4m ago

Granola transcribes computer audio locally

Granola is a local-first AI meeting notepad that transcribes audio directly on the user's device without using intrusive bots. It integrates automated transcription with manual note-taking to capture key insights while preserving meeting privacy.

BENCHMARK1h ago

Grok 4.5 ranks second on FrontierSWE

SpaceXAI's new Grok 4.5 model ranks second overall on Proximal's FrontierSWE software engineering benchmark and first in research capabilities. Trained in collaboration with Cursor, the model features a 500,000-token context window optimized for reasoning and agentic tasks.

UPDATE1h ago

@agent-browser/eve adds browser automation to Eve

The `@agent-browser/eve` package enables Vercel's filesystem-first agent framework, `eve`, to run browser automation tasks through `agent-browser`. By installing the `@agent-browser/eve` package, developers can equip their AI agents with a browser to fill out forms, test websites, collect data, and automate workflows.