NVIDIA Nemotron 3 Super 120B A12B BF16 REAP 50pct Draft is an open-source language model by 0xSero. Features: 120b LLM, VRAM: 128.5GB, Context: 256K, License: other, LLM Explorer Score: 0.31.
Arxiv:2510.13999 Base model:finetune:nvidia/nvi... Base model:nvidia/nvidia-nemot... Conversational Custom code Draft Endpoints compatible Latent-moe Nemotron h Pruned Reap Region:us Safetensors Sharded Sparse-moe Tensorflow
Model Card on HF 🤗: https://huggingface.co/0xSero/NVIDIA-Nemotron-3-Super-120B-A12B-BF16-REAP-50pct-draft
NVIDIA Nemotron 3 Super 120B A12B BF16 REAP 50pct Draft Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
NVIDIA Nemotron 3 Super 120B A12B BF16 REAP 50pct Draft Parameters and Internals
| LLM Name | NVIDIA Nemotron 3 Super 120B A12B BF16 REAP 50pct Draft |
| Repository 🤗 | https://huggingface.co/0xSero/NVIDIA-Nemotron-3-Super-120B-A12B-BF16-REAP-50pct-draft |
| Base Model(s) | |
| Model Size | 120b |
| Required VRAM | 128.5 GB |
| Updated | 2026-03-28 |
| Maintainer | 0xSero |
| Model Type | nemotron_h |
| Model Files | |
| Model Architecture | NemotronHForCausalLM |
| License | other |
| Context Length | 262144 |
| Model Max Length | 262144 |
| Transformers Version | 5.2.0.dev0 |
| Tokenizer Class | PreTrainedTokenizerFast |
| Padding Token | <|im_end|> |
| Vocabulary Size | 131072 |
Best Alternatives to NVIDIA Nemotron 3 Super 120B A12B BF16 REAP 50pct Draft
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| ...on 3 Super 120B A12B Base BF16 | 1024K / 209.5 GB | 10442 | 25 |
| ...motron 3 Super 120B A12B NVFP4 | 256K / 80.3 GB | 1058108 | 219 |
| ...Nemotron 3 Super 120B A12B FP8 | 256K / 128.4 GB | 873745 | 202 |
| ...emotron 3 Super 120B A12B BF16 | 256K / 194.6 GB | 163028 | 308 |
| ...motron 3 Super 120B A12B NVFP4 | 256K / 80.3 GB | 46082 | 22 |
| ...Nemotron 3 Super 120B A12B FP8 | 256K / 128.4 GB | 2514 | 9 |
| ...uper 120B A12B BF16 Heretic V2 | 256K / 241.4 GB | 248 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than 0xSero/NVIDIA-Nemotron-3-Super-120B-A12B-BF16-REAP-50pct-draft .
Rank the NVIDIA Nemotron 3 Super 120B A12B BF16 REAP 50pct Draft Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 52221 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a