NVIDIA
Nemotron 3.5 Lightning 30B-A3B
MoE
31,6B
Parameters · 3B active
23,7 Go
Q4_K_M file
≥ 25,2 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 25,2 Go (file + 4,000-token context). An estimate, not a measurement. Click a card for the detailed estimate.
Technical sheet
- Published
- 2026-08-01
- Architecture
- 52 layers · 2 KV heads · dimension 128
- Attention
- Hybrid: 6 of 52 layers use standard attention, the others a fixed-memory mechanism (Mamba, linear attention) — a very light cache.
- Context cache (f16)
- ≈ 6 MB per 1,000 tokens · 0,2 Go for 32,768 tokens
- GGUF file
- bartowski/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-GGUF · NVIDIA-Nemotron-3.5-Lightning-30B-A3B-Q4_K_M.gguf Download (23,7 Go)
- Official repo
- nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16
Public Hugging Face data (official config.json, GGUF repo). The context cache only counts full-attention layers.