# lumi-4b

> O Overall é a média aritmética simples dessas cinco métricas, com peso igual para cada uma. Nesse recorte, o Lumi-4B ocupa a 2ª posição entre 9 modelos.

- **Organisation:** ceia-br
- **Category:** Chat & Text Generation
- **Parameters:** 4B
- **Context length:** 262,144 tokens
- **Licence:** Unknown licence (Check licence)
- **Released:** 2026-10-27
- **Hugging Face:** 4.1K downloads (30 days), 75 likes

## Estimated memory

| Quant | ≈ GB |
|---|---|
| Q2_K | 3.09 |
| Q3_K_M | 3.71 |
| Q4_K_M | 4.17 |
| Q5_K_M | 4.58 |
| Q6_K | 5.01 |
| Q8_0 | 5.94 |
| F16 | 9.58 |

## Runs with

- [llama.cpp](https://selfhost.directory/project/llamacpp)
- [SGLang](https://selfhost.directory/project/sglang)
- [vLLM](https://selfhost.directory/project/vllm)

## About

O Overall é a média aritmética simples dessas cinco métricas, com peso igual para cada uma. Nesse recorte, o Lumi-4B ocupa a 2ª posição entre 9 modelos.

## Alternatives

- [nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16](https://selfhost.directory/models/nvidia/nvidia-nemotron-3-nano-4b-bf16) (4B, nvidia-nemotron-open-model-license, ≈4.17 GB at Q4): NVIDIA-Nemotron-3-Nano-4B-BF16 is a small language model (SLM) trained from scratch by NVIDIA, and designed as a unified model for both reasoning and non-reasoning tasks. It responds to user queries and tasks by first generating a reasoning trace and then concluding with a final response.
- [Qwen/Qwen3-4B-Instruct-2507](https://selfhost.directory/models/qwen/qwen3-4b-instruct-2507) (4B, Apache 2.0, ≈4.01 GB at Q4): We introduce the updated version of the Qwen3-4B non-thinking mode, named Qwen3-4B-Instruct-2507, featuring the following key enhancements:
- [Qwen/Qwen3-4B](https://selfhost.directory/models/qwen/qwen3-4b) (4B, Apache 2.0, ≈4.01 GB at Q4): Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.
- [Nanbeige/Nanbeige4.2-3B](https://selfhost.directory/models/nanbeige/nanbeige4.2-3b) (4.2B, Apache 2.0, ≈3.66 GB at Q4): Nanbeige4.2-3B is a compact agentic model built on Nanbeige4.2-3B-Base, designed to combine strong agentic behavior with broad reasoning and alignment capabilities. Its Looped Transformer architecture reuses the transformer layers to increase model capacity without adding parameters.
- [ibm-granite/granite-4.1-3b](https://selfhost.directory/models/ibm-granite/granite-4.1-3b) (3.4B, Apache 2.0, ≈3.14 GB at Q4): Model Summary: Granite-4.1-3B is a 3B parameter long-context instruct model finetuned from Granite-4.1-3B-Base using a combination of open source instruction datasets with permissive license and internally collected synthetic datasets.
- [XHToken/Spark-X2.6-4B](https://selfhost.directory/models/xhtoken/spark-x2.6-4b) (4.1B, Apache 2.0, ≈4.06 GB at Q4): A little over a month has passed since Spark-X2.5. In the rapidly evolving world of AI agents, that is enough time to rethink, refine, and move the frontier forward.

More: https://selfhost.directory/models/ceia-br/lumi-4b/alternatives

## Runs well on

- [GeForce RTX 4060](https://selfhost.directory/models/ceia-br/lumi-4b/on/rtx-4060): Q8_0, est. 35 tok/s
- [Apple M4 (16 GB)](https://selfhost.directory/models/ceia-br/lumi-4b/on/apple-m4-16gb): F16, est. 8.3 tok/s
- [Radeon RX 6700 XT](https://selfhost.directory/models/ceia-br/lumi-4b/on/rx-6700-xt): F16, est. 27 tok/s
- [GeForce RTX 3060 12GB](https://selfhost.directory/models/ceia-br/lumi-4b/on/rtx-3060): F16, est. 25 tok/s
- [Arc B580](https://selfhost.directory/models/ceia-br/lumi-4b/on/arc-b580): F16, est. 32 tok/s
- [Tesla T4](https://selfhost.directory/models/ceia-br/lumi-4b/on/tesla-t4): F16, est. 22 tok/s

## Similar size

- [meta-llama/Llama-3.2-3B-Instruct](https://selfhost.directory/models/meta-llama/llama-3.2-3b-instruct) (3.2B)
- [Qwen/Qwen3-4B](https://selfhost.directory/models/qwen/qwen3-4b) (4B)
- [Qwen/Qwen2.5-3B-Instruct](https://selfhost.directory/models/qwen/qwen2.5-3b-instruct) (3.1B)
- [microsoft/Phi-4-mini-instruct](https://selfhost.directory/models/microsoft/phi-4-mini-instruct) (3.8B)
- [Qwen/Qwen3-4B-Instruct-2507](https://selfhost.directory/models/qwen/qwen3-4b-instruct-2507) (4B)
- [XHToken/Spark-X2.6-4B](https://selfhost.directory/models/xhtoken/spark-x2.6-4b) (4.1B)

## FAQ

**How much VRAM does lumi-4b need?**
About 4.2 GB at Q4_K_M with an 8K context. Q8_0 needs about 5.9 GB and full precision about 9.6 GB. Longer contexts need more memory for the KV cache.

**Can lumi-4b run on the RTX 4060?**
Yes. It fits at Q8_0 (about 5.9 GB), at an estimated 35 tokens/s. It fits comfortably on the Apple M4 (16 GB), Radeon RX 6700 XT, RTX 3060 12GB.

**Can I use lumi-4b commercially?**
Unclear. It is released under Unknown licence. No licence stated: assume all rights reserved. Check the licence on the model card before using it in a product.

**What is the context length of lumi-4b?**
262,144 tokens (256K). The full window needs far more memory than an 8K context.

**How do I run lumi-4b with llama.cpp?**
Run llama-server -hf Mateus-Nery/lumi-4b-GGUF:Q4_K_XL. It downloads the GGUF and starts an OpenAI-compatible server.

## Links

- huggingface: https://huggingface.co/ceia-br/lumi-4b

---
Source: https://selfhost.directory/models/ceia-br/lumi-4b