NVIDIA's Nemotron 3.5 Bolt E4B Runs in 4GB

Episode 1 · NVIDIA open-sources 4GB edge model Nemotron 3.5 Bolt E4B (2026-10-23, 11 posts)

On October 23, NVIDIA released an open-source small model, Nemotron 3.5 Bolt E4B, on Hugging Face—a conversational agent that can run locally with just 4GB of memory. The model is built on Google Gemma 4 E4B (per a repost, 8B parameters with 4B effective parameters, capable of running locally on edge devices like 6GB-memory phones) via Nemotron post-training. It preserves Gemma's conversational ability while strengthening multi-turn tool calling, instruction following, and coding, and adds vision support, targeting local voice/device assistants, interactive characters, and robotics.

Confirmed

• The model continues a fully open policy: weights, data, and training recipes are all freely available and customizable.
• The official performance chart shows Bolt E4B in the "fast and smart" zone on a scatter plot of representative benchmark average score vs. time to first token (TTFT), scoring 70.2 with roughly 320ms TTFT, outperforming the previous generation.
• NVIDIA cited ScaleAILabs statistics showing the model tops the MultiChallenge leaderboard at 56.03%, ahead of Gemma 4 E4B (the official post mentioned the latter sits around the 43% tier).

Why it matters

• In the on-device small-model race, NVIDIA leveraged the Gemma base to quickly ship an open model optimized for agent scenarios, differentiating through low latency, low memory footprint, and a fully open recipe (data + recipe, not just weights), directly targeting local agent deployment on phones and other edge devices.

• NVIDIA releases Nemotron 3.5 Bolt E4B, a 4GB open model for on-device conversational agents — NVIDIAAI · 2026-10-23
• NVIDIA's chart: Nemotron Bolt E4B tops small models at ~320ms first-token latency — NVIDIAAI · 2026-10-23
• NVIDIA releases open Nemotron 3.5 Bolt E4B 4GB with weights, data and recipes — NVIDIAAI · 2026-10-23
• NVIDIA ships Nemotron 3.5 Bolt E4B, an open vision model that runs on 6GB RAM — danielhanchen · 2026-10-23
• NVIDIA: Bolt E4B tops MultiChallenge at 56.03% with weights, data, recipes open — NVIDIAAI · 2026-10-24
• NVIDIA ships Nemotron-3.5-Bolt-E4B, a 4GB Gemma-4-derived SLM for low-latency on-device voice assistants — jacek2023 · 2026-10-24
• NVIDIA ships Nemotron 3.5 Bolt E4B, an 8B/4B-effective model beating Gemma-4 and Qwen3.5 — mervenoyann · 2026-10-24
• NVIDIA ships 4GB Nemotron 3.5 Bolt that runs locally on Jetson Orin Nano 8GB, eyed for Reachy mini robots — alexcovo_eth · 2026-10-24
• NVIDIA ships 4GB Nemotron 3.5 Bolt E4B, a small open model that chats and calls tools on an 8GB Jetson — MaziyarPanahi · 2026-10-24
• NVIDIA ships Nemotron 3.5 Bolt E4B, a 4GB open model for robots and on-device agents — AIFlow_ML · 2026-10-24
• NVIDIA publishes benchmarks where its own model loses: Tau2 55.1 vs Qwen 3.5 4B's 75.1 — AIFlow_ML · 2026-10-24

Episode 2 · NVIDIA Nemotron 3.5 Bolt Arrives on Ollama for Local Deployment (2026-10-24, 3 posts)

NVIDIA's Nemotron 3.5 Bolt E4B is now available on Ollama as a small open vision-language model with 128K context, image understanding, and tool calling, runnable locally with as little as 4GB.

• NVIDIA's Nemotron 3.5 Bolt lands on Ollama: small VLM with 128K context — ollama · 2026-10-24
• NVIDIA's Nemotron 3.5 Bolt lands on Ollama: local vision model with 128K context — ollama · 2026-10-24
• NVIDIA's Nemotron 3.5 Bolt E4B small open model lands on Ollama for local use — ollama · 2026-10-24

Episode 3 · 4GB Nemotron Bolt E4B Runs Robots on Jetson (2026-10-25, 2 posts)

Developers demonstrated NVIDIA's Nemotron 3.5 Bolt E4B, a 4GB model built on Gemma 4 E4B with Nemotron post-training, running locally on Jetson hardware to drive a Reachy Mini robot.

• What can a 4GB model do on a robot? Nemotron 3.5 Bolt E4B tested on Jetson — lmoroney · 2026-10-25
• What a 4GB model can do on a robot: Nemotron Bolt runs locally on Reachy Mini — lmoroney · 2026-10-25