The AI Loadout
←Back to models
Checked Active: 2026-10-25
NVIDIA

Nemotron 3.5 Bolt E4B (local)

Fast local voice and device assistants, game characters and small tool-using agents on a 16 GB laptop, Mac or PC.

Evaluation Analysis: Released Oct 23, 2026 and added to Ollama in v0.40.10 the same day. NVIDIA post-trained Gemma 4 E4B for conversation and multi-turn tool calls: 8B parameters, about 4B effective. Reasoning can be switched on or off, and a conversational mode keeps replies short and speech-ready. Takes images in Ollama. Ships with a DFlash draft for speculative decoding, which NVIDIA says gives up to 21% more throughput than Gemma 4 E4B's own drafter. NVIDIA reports about 40% higher average scores than its earlier Nemotron 3 Nano 4B on its conversational agent benchmarks. Unlike the 74B-A9B Bolt, the E4B runs on Apple Silicon as well as Windows and Linux.
Context Window128K (64K for the MLX tags)
Input Price / 1MFree (local)
Output Price / 1MFree (local)
Release Date2026-10-23

Run it locally

ollama run nemotron-3.5-bolt:e4b
Ollama tag
nemotron-3.5-bolt:e4b (also e4b-q4_K_M, e4b-q8_0, e4b-bf16 and MLX tags)
LM Studio name
nvidia/nemotron-3.5-bolt-e4b (5 GB minimum memory)
Download size by quantization
e4b 7.2 to 7.9 GB; Q8_0 9.8 GB; BF16 17 GB
License
OpenMDW 1.1

Memory needed is roughly the download size plus 2 to 4 GB for context. See how to run AI locally.

Specs and prices checked 2026-10-25 against the official page.

Used in these setups

Related skills and personas

Agent skills and bot personas that pair well with Nemotron 3.5 Bolt E4B (local).

Browse full directories:All Skills (>35)·All Personas (>15)

Want tested skills and personas installed in one command? See the AI Loadout packs, from $9.