NVIDIA Nemotron 3 Super 120B A12B FP8 — Hardware Requirements & GPU Compatibility
Chat
NVIDIA Nemotron 3 Super 120B A12B FP8 is the FP8 variant of NVIDIA's largest Nemotron 3 mixture-of-experts model, weighing in at 123.6 billion parameters. With 12 billion parameters active per token, it delivers exceptional reasoning and conversational depth while the FP8 format keeps memory usage lower than full precision. This model sits at the high end of what's achievable for local inference. You'll need serious GPU memory to run it, but the payoff is near-frontier model quality running entirely on your own hardware. The FP8 quantization offers a meaningful memory reduction over BF16 with minimal quality trade-off.
64.6K downloads
132 likesSpecifications
- Publisher
- NVIDIA
- Parameters
- 123.6B
- Architecture
- NemotronHForCausalLM
- Context Length
- 262,144 tokens
- Vocabulary Size
- 131,072
- Release Date
- 2026-03-14
- License
- Other
Get Started
HuggingFace
How Much VRAM Does NVIDIA Nemotron 3 Super 120B A12B FP8 Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16 | 16.00 | 247.7 GB | 271.1 GB | 247.22 GB | Brain floating point 16 — preferred for training |
Which GPUs Can Run NVIDIA Nemotron 3 Super 120B A12B FP8?
BF16 · 247.7 GBNVIDIA Nemotron 3 Super 120B A12B FP8 (BF16) requires 247.7 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 323+ GB is recommended. Using the full 262K context window can add up to 23.4 GB, bringing total usage to 271.1 GB. No single GPU has enough memory — multi-GPU or cluster setups are needed.
Which Devices Can Run NVIDIA Nemotron 3 Super 120B A12B FP8?
BF16 · 247.7 GB2 devices with unified memory can run NVIDIA Nemotron 3 Super 120B A12B FP8, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomRelated Models
Frequently Asked Questions
- How much VRAM does NVIDIA Nemotron 3 Super 120B A12B FP8 need?
NVIDIA Nemotron 3 Super 120B A12B FP8 requires 247.7 GB of VRAM at BF16. Full 262K context adds up to 23.4 GB (271.1 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 123.6B × 16 bits ÷ 8 = 247.2 GB
KV Cache + Overhead ≈ 0.5 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 23.9 GB (at full 262K context)
VRAM usage by quantization
BF16247.7 GBBF16 + full context271.1 GB- Can NVIDIA GeForce RTX 5090 run NVIDIA Nemotron 3 Super 120B A12B FP8?
No — NVIDIA Nemotron 3 Super 120B A12B FP8 requires at least 247.7 GB at BF16, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
- Can I run NVIDIA Nemotron 3 Super 120B A12B FP8 on a Mac?
NVIDIA Nemotron 3 Super 120B A12B FP8 requires at least 247.7 GB at BF16, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run NVIDIA Nemotron 3 Super 120B A12B FP8 locally?
Yes — NVIDIA Nemotron 3 Super 120B A12B FP8 can run locally on consumer hardware. At BF16 quantization it needs 247.7 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- What's the download size of NVIDIA Nemotron 3 Super 120B A12B FP8?
At BF16, the download is about 247.22 GB.