Nemotron 3.5 Bolt (74B-A9B, 2-bit)
NVIDIA's Nemotron 3.5 Bolt: Super VL distilled down to 74B parameters with 9B active, quantization-aware trained to fit in about 24GB. Native reasoning, tool calling, image understanding, and a 256K context window, with a small LiLiCorr drafter that speeds up replies. Needs an NVIDIA or Vulkan GPU with about 26GB of memory; not offered on Macs.
| At a glance | |
|---|---|
| Made by | NVIDIA |
| Download size | 22.0 GB |
| Parameters | 74B |
| Quantization | IQ2_XXS |
| Context window | 262K tokens |
| Uses tools | Yes |
| Best for | Reasoning |
| Runs on | llama.cpp |
| License | OpenMDW 1.1 |
| Added to the catalog | 2026-10-17 |
Gezel downloads this model the first time a gezel needs it, checks the file against its published hash, and applies tuned settings for it.