nvidia/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-BF16
main boardmodel
NVIDIA Nemotron 3.5 Bolt 74B-A9B is a vision-language model (VLM) designed to run everyday agentic tasks locally. Mixture-of-experts, 75B total, 9B active (512 experts). Context: 256K tokens (config). License: openmdw-1.1.
- Org
- nvidia
- Released
- 2026-10-09
- Status
- main board
- License
- openmdw-1.1 (permissive)
- Parameters
- 74.9B total, 9.0B active
- Architecture
- MoE · 512 experts, None/token · NemotronH_Omni_Reasoning_V3
- Context
- 262,144 tokens
- Task
- image-text-to-text · vision
- Signal
- 47 HF likes · 4731 downloads · 1 X post(s)
Runs on
real quant files; need = weights + FP16 KV cache for 8K tokens + 1GB. A sizing calculation, not tested.
- Q4 ~57.6 GB: Mac 96GB
- Q8 ~86.3 GB: 96GB GPU (RTX Pro 6000 Blackwell), 1× DGX Spark (128GB), Strix Halo 128GB unified (~96GB usable), Mac 128GB, Mac 192GB, CPU only, 128GB RAM (slow)
- FP16 ~155.6 GB: 2× DGX Spark (256GB), Mac 256GB, Mac 512GB, CPU only, 256GB RAM (slow)
Links
- Hugging Face
- Ollama
- GGUF: nvidia/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB-GGUF
- GGUF: unsloth/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-GGUF
First published here 2026-10-09 · updated 2026-10-09 · markdown · JSON