What are the hardware requirements for NVIDIA Nemotron Labs 3 Puzzle 75B A9B BF16?

NVIDIA Nemotron Labs 3 Puzzle 75B A9B BF16 requires approximately 150.8 GB of VRAM and supports a context window of 256K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed NVIDIA Nemotron Labs 3 Puzzle 75B A9B BF16 and how large is it?

NVIDIA Nemotron Labs 3 Puzzle 75B A9B BF16 is developed by nvidia, a model with 120b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate NVIDIA Nemotron Labs 3 Puzzle 75B A9B BF16?

NVIDIA Nemotron Labs 3 Puzzle 75B A9B BF16 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

NVIDIA Nemotron Labs 3 Puzzle 75B A9B BF16 by nvidia — VRAM 150.8GB, 256K context

Name: NVIDIA Nemotron Labs 3 Puzzle 75B A9B BF16
Author: nvidia

NVIDIA Nemotron Labs 3 Puzzle 75B A9B BF16 is an open-source language model by nvidia. Features: 120b LLM, VRAM: 150.8GB, Context: 256K, License: other, LLM Explorer Score: 0.3.

Arxiv:2411.19146 Arxiv:2512.20848 Arxiv:2604.12374 Arxiv:2607.04371 Base model:finetune:nvidia/nvi... Base model:nvidia/nvidia-nemot... Conversational Custom code Dataset:nvidia/nemotron-post-t... Dataset:nvidia/nemotron-pre-tr... De En Es Fr It Ja Latent-moe Mtp Nemotron-3 Nemotron h puzzle Nvidia Pytorch Region:us Safetensors Sharded Tensorflow Zh

Model Card on HF 🤗: https://huggingface.co/nvidia/NVIDIA-Nemotron-Labs-3-Puzzle-75B-A9B-BF16

NVIDIA Nemotron Labs 3 Puzzle 75B A9B BF16 Parameters and Internals

LLM Name	NVIDIA Nemotron Labs 3 Puzzle 75B A9B BF16
Repository 🤗	https://huggingface.co/nvidia/NVIDIA-Nemotron-Labs-3-Puzzle-75B-A9B-BF16
Base Model(s)	...emotron 3 Super 120B A12B BF16 nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16
Model Size	120b
Required VRAM	150.8 GB
Updated	2026-08-08
Maintainer	nvidia
Model Type	nemotron_h_puzzle
Model Files	5.0 GB: 1-of-31 5.0 GB: 2-of-31 5.0 GB: 3-of-31 5.0 GB: 4-of-31 5.0 GB: 5-of-31 5.0 GB: 6-of-31 5.0 GB: 7-of-31 5.0 GB: 8-of-31 5.0 GB: 9-of-31 5.0 GB: 10-of-31 5.0 GB: 11-of-31 4.9 GB: 12-of-31 5.0 GB: 13-of-31 5.0 GB: 14-of-31 5.0 GB: 15-of-31 5.0 GB: 16-of-31 5.0 GB: 17-of-31 5.0 GB: 18-of-31 5.0 GB: 19-of-31 5.0 GB: 20-of-31 5.0 GB: 21-of-31 5.0 GB: 22-of-31 5.0 GB: 23-of-31 4.9 GB: 24-of-31 5.0 GB: 25-of-31 5.0 GB: 26-of-31 5.0 GB: 27-of-31 5.0 GB: 28-of-31 5.0 GB: 29-of-31 4.9 GB: 30-of-31 1.1 GB: 31-of-31 5.9 GB
Supported Languages	en fr es it de ja zh
Model Architecture	NemotronHPuzzleForCausalLM
License	other
Context Length	262144
Model Max Length	262144
Transformers Version	4.56.0
Tokenizer Class	PreTrainedTokenizerFast
Padding Token	<\|im_end\|>
Vocabulary Size	131072

Rank the NVIDIA Nemotron Labs 3 Puzzle 75B A9B BF16 capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to NVIDIA Nemotron Labs 3 Puzzle 75B A9B BF16

Best Alternatives	Context / RAM	Downloads	Likes
...on Labs 3 Puzzle 75B A9B NVFP4	256K / 47.6 GB	164224	128
...tron Labs 3 Puzzle 75B A9B FP8	256K / 77.1 GB	23607	17

Note: green Score (e.g. "73.2") means that the model is better than nvidia/NVIDIA-Nemotron-Labs-3-Puzzle-75B-A9B-BF16.