Wirestack
NVIDIA Nemotron 3.5 Super and Bolt run in Ollama from v0.40.5 (October 14, 2026)
A 120B vision-language model (12B active, 1M context) and a 74B Bolt version that fits a 24 GB NVIDIA GPU, local in Ollama.
3 of 3 free copies left this month.Unlimited with Pro
What you get
Ollama v0.40.5 (October 14, 2026) adds NVIDIA Nemotron 3.5 Super, a vision-language model for coding, reasoning, tool use and documents. It is a 120B-parameter mixture-of-experts model with 12B active per token, tool calling, thinking you can switch on or off, and up to 1M tokens of context. Nemotron 3.5 Bolt (74B, 9B active) is tuned to fit a 24 GB NVIDIA GPU.
Prerequisites
- Ollama v0.40.5 or newer
- No API key for local use
Steps
- Update Ollama to v0.40.5 or newer.
- Run
ollama run nemotron-3.5-super.
Test it
Ask the model a question in the terminal and check it answers.
Not confirmed
- How much memory Super needs on your own machine; check the model page before downloading.