Runs locally on a Mac · by NVIDIA
Nemotron Bolt
Run a small coding and tool-using helper on your Mac.
text, image, speech → text, data, code
- Way in
- Official APINo sign-up, no key
- To start
- Freeopen weights (OpenMDW-1.1)
- Runs
- On your MacAlso through Ollama
- Checked
- Nov 3, 2026Prices as of that day
What it can do 1
Local coding/tool-use assistant in the 4B-effective classPull the explicit tag nemotron-3.5-bolt:e4b (q4_K_M 7.9 GB, nvfp4 MLX 7.2 GB): the :latest tag is the 74B-A9B model (25 GB) and will not fit in 16 GB. HF ‘4GB’ GGUF; 64K in/out per HF card (Ollama GGUF tags list 128K).text, image, speech → text, data, codefree (local)
Prices and limits as of Nov 3, 2026, from NVIDIA’s own pages. They change; check before you rely on them.Not affiliated with NVIDIA. Not advice.Something wrong?