What are the hardware requirements for Nemotron 3.5 Lightning 30B A3B REAP 20B?

Nemotron 3.5 Lightning 30B A3B REAP 20B requires approximately 39.7 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 Nemotron 3.5 Lightning 30B A3B REAP 20B and how large is it?

Nemotron 3.5 Lightning 30B A3B REAP 20B is developed by sleepyeldrazi, a model with 30b 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 Nemotron 3.5 Lightning 30B A3B REAP 20B?

Nemotron 3.5 Lightning 30B A3B REAP 20B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Nemotron 3.5 Lightning 30B A3B REAP 20B by sleepyeldrazi — VRAM 39.7GB, 256K context

Name: Nemotron 3.5 Lightning 30B A3B REAP 20B
Author: sleepyeldrazi

Nemotron 3.5 Lightning 30B A3B REAP 20B is an open-source language model by sleepyeldrazi. Features: 30b LLM, VRAM: 39.7GB, Context: 256K, License: openmdw-1.1, LLM Explorer Score: 0.31.

Arxiv:2510.13999 Base model:finetune:nvidia/nvi... Base model:nvidia/nvidia-nemot... Conversational Endpoints compatible Experimental Expert-pruning Hybrid Mamba2 Moe Nemotron-3.5 Nemotron h Nvidia Reap Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/sleepyeldrazi/Nemotron-3.5-Lightning-30B-A3B-REAP-20B

Nemotron 3.5 Lightning 30B A3B REAP 20B Parameters and Internals

LLM Name	Nemotron 3.5 Lightning 30B A3B REAP 20B
Repository 🤗	https://huggingface.co/sleepyeldrazi/Nemotron-3.5-Lightning-30B-A3B-REAP-20B
Base Model(s)	nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16
Model Size	30b
Required VRAM	39.7 GB
Updated	2026-08-14
Maintainer	sleepyeldrazi
Model Type	nemotron_h
Model Files	20.0 GB: 1-of-2 19.7 GB: 2-of-2
Model Architecture	NemotronHForCausalLM
License	openmdw-1.1
Context Length	262144
Model Max Length	262144
Transformers Version	5.10.1
Tokenizer Class	TokenizersBackend
Padding Token	<\|im_end\|>
Vocabulary Size	131072

Rank the Nemotron 3.5 Lightning 30B A3B REAP 20B 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 Nemotron 3.5 Lightning 30B A3B REAP 20B

Best Alternatives	Context / RAM	Downloads	Likes
...on 3.5 Lightning 30B A3B NVFP4	1024K / 12.6 GB	19250	100
...30B A3B NVFP4 Global Pruned 15	1024K / 11 GB	148	19
...A Nemotron 3 Nano 30B A3B BF16	256K / 63.2 GB	1645889	752
...ron 3.5 Lightning 30B A3B BF16	256K / 65.9 GB	15740	62
... Nemotron 3 Nano 30B A3B NVFP4	256K / 19.3 GB	1568	8
Nemotron Cascade 2 30B A3B	256K / 63.2 GB	88240	518
...IA Nemotron 3 Nano 30B A3B FP8	256K / 32.7 GB	566152	356
... Nemotron 3 Nano 30B A3B NVFP4	256K / 19.3 GB	765499	174
...Nemotron 3.5 Lightning 30B A3B	256K / 65.9 GB	474	4
....5 Lightning 30B A3B Base BF16	256K / 65.9 GB	336	12

Note: green Score (e.g. "73.2") means that the model is better than sleepyeldrazi/Nemotron-3.5-Lightning-30B-A3B-REAP-20B.