--- format: "aidr-story-markdown/v1" id: "7dd37c43bde87d3c18064849fd65352d4991a96a866fb6cc5df77c2cdc6ff492" canonical_url: "https://aidr.today/7dd37c43?lang=en" title: "Nvidia Releases Nemotron 3.5 Bolt E4B Small Open Model for Edge AI" lang: "en" requested_lang: "en" available_langs: ["en","vi"] translation_fallback: null fallback_fields: [] published_at: "2026-10-23T16:08:14.000Z" category: "Models" topics: ["nvidia","llm"] source_urls: ["https://huggingnews.com/ai/nvidia-releases-nemotron-35-bolt-e4b-small-open-model-for-edge-ai-5c43d721","https://x.com/NVIDIAAI/status/2113657112408785153","https://x.com/UnslothAI/status/2113658964365971681","https://marketbrief.now/ai/nvidia-releases-nemotron-35-bolt-e4b-small-open-model-for-edge-ai-5c43d721"] summary: "Nvidia released the Nemotron 3.5 Bolt E4B Small open model, optimized for mobile phones and robotics with limited memory. The model supports vision and tool use, and was demoed on the Jetson Orin Nano 8GB alongside the NVIDIA Cosmos vision model to power a Reachy Mini robot. While Nvidia lists the model as 4GB, Unsloth states it runs locally on 6GB RAM." --- # Nvidia Releases Nemotron 3\.5 Bolt E4B Small Open Model for Edge AI > [Open the canonical story]() **Published:** 2026-10-23T16:08:14.000Z **Category:** Models **Topics:** nvidia, llm ## Summary Nvidia released the Nemotron 3\.5 Bolt E4B Small open model, optimized for mobile phones and robotics with limited memory\. The model supports vision and tool use, and was demoed on the Jetson Orin Nano 8GB alongside the NVIDIA Cosmos vision model to power a Reachy Mini robot\. While Nvidia lists the model as 4GB, Unsloth states it runs locally on 6GB RAM\. ## Sources - [Story source]() - [Story source]() - [Supporting source]() - [Story source]()