NVIDIA
Nemotron 3 Super 120B-A12B
MoE
123,6B
Parameters · 12B active
78,0 Go
Q4_K_XL file
≥ 79,5 Go
Recommended VRAM
262 144
Max context
Not measured yet
No card has run the TokenBench protocol on this model yet. The current protocol covers a selection of models; benchmarking any catalog model from the app is coming soon.
Measure my card →Which cards should it fit on?
Cards on the site whose video memory covers the recommended 79,5 Go (file + 4,000-token context). An estimate, not a measurement. Click a card for the detailed estimate.
Technical sheet
- Published
- 2026-03-10
- Architecture
- 88 layers · 2 KV heads · dimension 128
- Attention
- Hybrid: 8 of 88 layers use standard attention, the others a fixed-memory mechanism (Mamba, linear attention) — a very light cache.
- Context cache (f16)
- ≈ 8 MB per 1,000 tokens · 0,3 Go for 32,768 tokens
- GGUF file
- unsloth/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF · 3 files Download (78,0 Go)
- Official repo
- nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16
Public Hugging Face data (official config.json, GGUF repo). The context cache only counts full-attention layers.