What are the hardware requirements for Lumi 4B?

Lumi 4B requires approximately 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 Lumi 4B and how large is it?

Lumi 4B is developed by ceia-research, a model with 4b 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 Lumi 4B?

Lumi 4B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Lumi 4B by ceia-research — VRAM 8GB, 256K context

Name: Lumi 4B
Author: ceia-research

Lumi 4B is an open-source language model by ceia-research. Features: 4b LLM, VRAM: 8GB, Context: 256K, LLM Explorer Score: 0.4.

Base model:finetune:nvidia/nvi... Base model:nvidia/nvidia-nemot... Continual-pretraining Conversational Custom code Dataset:huggingfacefw/finepdfs... Dataset:huggingfacefw/fineweb-... Dataset:nvidia/nemotron-cc-v2.... Dataset:nvidia/nemotron-pretra... Dataset:nvidia/nemotron-pretra... Dataset:polygl0t/gigaverbo-v2 Knowledge-distillation Nemotron h Portuguese Pt Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/ceia-research/lumi-4b

Lumi 4B Parameters and Internals

LLM Name	Lumi 4B
Repository 🤗	https://huggingface.co/ceia-research/lumi-4b
Base Model(s)	NVIDIA Nemotron 3 Nano 4B BF16 nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16
Model Size	4b
Required VRAM	8 GB
Updated	2026-11-13
Maintainer	ceia-research
Model Type	nemotron_h
Model Files	8.0 GB
Supported Languages	pt
Model Architecture	NemotronHForCausalLM
Context Length	262144
Model Max Length	262144
Transformers Version	5.8.1
Tokenizer Class	TokenizersBackend
Vocabulary Size	131072

Rank the Lumi 4B 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 Lumi 4B

Best Alternatives	Context / RAM	Downloads	Likes
Lumi 4B	256K / 8 GB	191	19
...oppen Gen4k Ep2 Ncp5 N3nc Vs16	256K / 8 GB	2477	1
...ppen Gen4k Ep2 Ncp5 N3nc Iid16	256K / 8 GB	2447	1
...oppen Gen4k Ep2 Ncp5 N3nc Base	256K / 8 GB	2231	1
NVIDIA Nemotron 3 Nano 4B FP8	256K / 5.3 GB	83285	27
H2o Lightning 4B Nemotron Nano	256K / 8 GB	279	3
... Nano 4B 8K Grpo Hardyman S300	256K / 15.9 GB	338	1
NVIDIA Nemotron 3 Nano 4B	256K / 8 GB	4851	16
NVIDIA Nemotron 3 Nano 4B FP8	256K / 10.6 GB	343	5
NVIDIA Nemotron 3 Nano 4B BF16	256K / 7.9 GB	131	2