Ollama 0.40.10 adds Nemotron 3.5 Bolt E4B, a small Gemma 4-based vision model with tool use and a 128K context. Here’s a generic way to try it safely.
Review before you install. This model tag is brand new. Read its page on ollama.com first and check its size, license and publisher. Try it on a machine without sensitive data before you wire it into anything.
1. Update Ollama
Use your normal package manager or the official installer, then confirm the version:
ollama --version2. Pull the model and check its size
ollama pull nemotron-3.5-bolt:e4b
ollama list3. Chat with it
ollama run nemotron-3.5-bolt:e4b4. Try an image
It’s a vision model, so give it a picture with no personal info in it:
ollama run nemotron-3.5-bolt:e4b "Describe this image: ./test.png"5. Watch your resources
A 128K context uses a lot of memory. Check ollama ps while it runs, and keep the context smaller if your machine struggles.
6. Keep it local
By default Ollama listens on localhost. Don’t expose port 11434 to your network without a reason and some authentication in front of it.
Have a blessed day! And Happy Hacking!
