Bolt E4B is NVIDIA’s Gemma 4 E4B derivative, with 8B total and 4B effective parameters. Its local Ollama tag supports Windows/Linux and Apple silicon. Choose e4b explicitly: default Bolt is a different, larger model.
Where it works well
- Text and image input.
- Coding and reasoning assistance.
- Tool calling in compatible apps.
Common use cases
- Explain a small function.
- Ask about a screenshot.
- Draft a checklist from notes.
Know its limits
- E4B is an effective count, not a 4B memory footprint.
- The separate 74B-A9B Bolt currently lacks Apple silicon/Metal support.
- A DFlash draft model or images add memory beyond this short-text estimate.
Run it on your computer
Compatibility: Choose the E4B tag for Windows/Linux or Apple silicon.
Start with a compressed download and a short conversation. Our estimate is about 16 GB of total RAM or Mac unified memory, with room for the operating system. The listed 7.2–7.9 GB download is the model file, not its full running memory use.
Check the model’s supported runtime and hardware before loading it. System RAM on a PC and unified memory on a Mac are different from dedicated NVIDIA VRAM. A memory match doesn’t guarantee compatibility or speed. Choose a smaller model if responses are too slow.
Use a desktop app that supports the listed model format, such as LM Studio or Ollama. Download the exact local variant, load it in Chat, and start with one small task. Model downloads need internet; local chat can work offline once the model is downloaded.
Official model information
Model capabilities checked October 2026. The use cases above are practical starting suggestions, not performance guarantees.