Property	Value
Base model	0xSero/NVIDIA-Nemotron-3-Super-120B-A12B-BF16-REAP-50pct-draft
Upstream	nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16
Architecture	NemotronH (Mamba2 + Transformer hybrid MoE)
Active parameters / token	~12B
Quantization	INT4 weight-only
Approx. on-disk size	~39 GB
Languages	English

Load (vLLM)

python

from vllm import LLM, SamplingParams

llm = LLM(
    model="drawais/Nemotron-3-Super-120B-A12B-REAP50-W4A16",
    trust_remote_code=True,
    max_model_len=65536,
)
params = SamplingParams(temperature=0.6, top_p=0.95, max_tokens=4096)
print(llm.generate(["Hello, world!"], params)[0].outputs[0].text)

Serve (vLLM, OpenAI-compatible)

bash

vllm serve drawais/Nemotron-3-Super-120B-A12B-REAP50-W4A16 \
  --trust-remote-code \
  --max-model-len 65536 \
  --gpu-memory-utilization 0.94

Then use any OpenAI-compatible client against http://127.0.0.1:8000/v1. For reasoning-mode controls, recommended --reasoning-parser, tool-calling flags, and per-modality serving recommendations, follow the upstream Nemotron-3-Super model card.

Footprint

~39 GB on disk for the weights. Recommended VRAM:

1× 80 GB GPU (H100 / H200 / A100 80 GB) for full context
2× 48 GB GPUs with tensor parallelism for headroom

trust_remote_code=True is required.

Sibling formats

For the same REAP-50 source (by 0xSero):

0xSero/NVIDIA-Nemotron-3-Super-120B-A12B-BF16-REAP-50pct-draft — BF16 source
0xSero/NVIDIA-Nemotron-3-Super-120B-A12B-BF16-REAP-50pct-AutoRound-W4A16-draft — alternative INT4 (AutoRound)

Acknowledgements

REAP pruning courtesy of 0xSero/NVIDIA-Nemotron-3-Super-120B-A12B-BF16-REAP-50pct-draft. Source model © NVIDIA Corporation, released under the NVIDIA Open Model Agreement. This artifact is a Derivative Work as defined in that agreement. See LICENSE and NOTICE for full text and required attribution.

License

NVIDIA Open Model Agreement (Release Date: April 2, 2026).

Commercially usable. You are free to create and distribute Derivative Works. NVIDIA does not claim ownership of outputs.

The full agreement text is included in LICENSE. The attribution notice required by Section 3(c) is in NOTICE:

Licensed by NVIDIA Corporation under the NVIDIA Open Model Agreement.