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 to lumi-4b
120 open-weight models that do the same job as lumi-4b (4B chat), ranked by how close they are in task and size, then by popularity and how recent they are. Licence and VRAM are shown for each.
Model Summary: Granite‑4.1‑3B‑Base is a decoder‑only language model with long‑context capabilities, designed to support a broad range of general text‑to‑text generation tasks, as well as fill‑in‑the‑Middle (FIM) code completion.
3.4B128K ctx Apache 2.010.6K 26Apr 2026- Same category
- Similar size (3.4B vs 4B)
- Commercial use OK
Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2:
3.1B32K ctx Qwen Research License2M 641Sept 2024- Same category
- Similar size (3.1B vs 4B)
- Widely used
Jan-Code-4B is a code-tuned model built on top of Jan-v3-4B-base-instruct. It’s designed to be a practical coding model you can run locally and iterate on quickly—useful for everyday code tasks and as a lightweight “worker” model in agentic workflows.
4.4B256K ctx Apache 2.08K 83Mar 2026- Same category
- Similar size (4.4B vs 4B)
- Commercial use OK
Phi-3.5-mini is a lightweight, state-of-the-art open model built upon datasets used for Phi-3 - synthetic data and filtered publicly available websites - with a focus on very high-quality, reasoning dense data. The model belongs to the Phi-3 model family and supports 128K token context length.
3.8B128K ctx MIT340.3K 1.3KAug 2024- Same category
- Similar size (3.8B vs 4B)
- Commercial use OK
Raon-1 nano is a dense language model with 1.5B effective parameters (3.1B total, including a 1.5B per-layer embedding table), developed by KRAFTON AI. It is designed to follow instructions, maintain conversational context, and call the right tools in both Korean and English.
3.5B256K ctx Apache 2.01.3K 95Oct 2026- Same category
- Similar size (3.5B vs 4B)
- Commercial use OK
- Newer release
Phi-tiny-MoE is a lightweight Mixture of Experts (MoE) model with 3.8B total parameters and 1.1B activated parameters. It is compressed and distilled from the base model shared by Phi-3.5-MoE and GRIN-MoE using the SlimMoE approach, then post-trained via supervised fine-tuning and direct preference…
3.8BA1.3B4K ctx MIT30.3K 52Jun 2025- Same category
- Similar size (3.8B vs 4B)
- Commercial use OK
The Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out).
3.2B128K ctx Llama 3.2 Community License834.2K 3KSept 2024- Same category
- Similar size (3.2B vs 4B)
Model Summary: Granite Switch 4.1 3B Preview is a modular LLM built on IBM Granite 4.1 3B with embedded adapters from the Granite Libraries collection.
4.1B128K ctx Apache 2.01.8K 37May 2026- Same category
- Similar size (4.1B vs 4B)
- Commercial use OK
Agents‑A1 is a long-horizon agentic model from InternScience, built to scale heterogeneous agentic abilities across multiple domains including Long‑horizon Search, Engineering, Scientific Research, Instruction Following, and Tool-calling.
4.5B256K ctx Apache 2.02.3K 111Jul 2026- Same category
- Similar size (4.5B vs 4B)
- Commercial use OK
AREX is a family of deep research agents developed by the Beijing Academy of Artificial Intelligence (BAAI). It is designed for long-horizon tasks in which an agent must search across sources, assemble candidate answers, verify multiple constraints, and revise its research plan when the available…
4.5B256K ctx Apache 2.01.9K 53Jul 2026- Same category
- Similar size (4.5B vs 4B)
- Commercial use OK
Nanbeige4.2-3B-Base is a new base model in the Nanbeige4 series, trained from scratch with improvements to both model architecture and training data.
4.2B128K ctx Apache 2.0889 64Jul 2026- Same category
- Similar size (4.2B vs 4B)
- Commercial use OK
📣 Update [10-07-2025]: Added a default system prompt to the chat template to guide the model towards more professional, accurate, and safe responses.
3.4B128K ctx Apache 2.022.2K 274Sept 2025- Same category
- Similar size (3.4B vs 4B)
- Commercial use OK
Nemotron-Mini-4B-Instruct is a model for generating responses for roleplaying, retrieval augmented generation, and function calling. It is a small language model (SLM) optimized through distillation, pruning and quantization for speed and on-device deployment.
4B4K ctx NVIDIA Open Model License99.7K 192Sept 2024- Same category
- Similar size (4B vs 4B)
- Commercial use OK
We elaborate on the entire construction pipeline of AgentCPM-Explore on GitHub.
4B256K ctx Apache 2.02.4K 418Jan 2026- Same category
- Similar size (4B vs 4B)
- Commercial use OK
T5Gemma is a family of lightweight yet powerful encoder-decoder research models from Google. These models are created by adapting pretrained decoder-only models into a encoder-decoder.
5.6B8K ctx Gemma Terms of Use223.4K 10Jun 2025- Same category
- Similar size (5.6B vs 4B)
SmolLM3 is a 3B parameter language model designed to push the boundaries of small models. It supports 6 languages, advanced reasoning and long context. SmolLM3 is a fully open model that offers strong performance at the 3B–4B scale.
3.1B64K ctx Apache 2.088.2K 180Jun 2025- Same category
- Similar size (3.1B vs 4B)
- Commercial use OK
The Phi-3-Mini-4K-Instruct is a 3.8B parameters, lightweight, state-of-the-art open model trained with the Phi-3 datasets that includes both synthetic data and the filtered publicly available websites data with a focus on high-quality and reasoning dense properties.
3.8B4K ctx MIT310.3K 1.5KApr 2024- Same category
- Similar size (3.8B vs 4B)
- Commercial use OK
Aloha! 🌺 Today, we are releasing Ornith-1.0, a self-improving family of open-source models for agentic coding.
9B256K ctx MIT822.7K 571Jun 2026- Same category
- Larger (9B vs 4B)
- Commercial use OK
Cohere Labs Tiny Aya is an open weights research release of a pretrained 3.35 billion parameter model optimized for efficient, strong, and balanced multilingual representation across 70+ languages, including many lower-resourced ones.
3.3BCC BY-NC 4.06K 33Feb 2026- Same category
- Similar size (3.3B vs 4B)
The Phi-3-Mini-128K-Instruct is a 3.8 billion-parameter, lightweight, state-of-the-art open model trained using the Phi-3 datasets. This dataset includes both synthetic data and filtered publicly available website data, with an emphasis on high-quality and reasoning-dense properties.
3.8B128K ctx MIT240.5K 1.7KApr 2024- Same category
- Similar size (3.8B vs 4B)
- Commercial use OK
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.
8.2B40K ctx Apache 2.06.1M 2.2KApr 2025- Same category
- Larger (8.2B vs 4B)
- Commercial use OK
- Widely used
The Llama 3.2 collection of multilingual large language models (LLMs) is a collection of pretrained and instruction-tuned generative models in 1B and 3B sizes (text in/text out).
3.2B128K ctx Llama 3.2 Community License313.5K 1.1KSept 2024- Same category
- Similar size (3.2B vs 4B)
We are releasing MiniCPM5-2B, the second model in the MiniCPM5 series, following MiniCPM5-1B. It is a dense 2B Transformer that scales up the same training recipe, built for on-device, local deployment, and resource-constrained scenarios, reaching 2B-class open-source SOTA.
2.5B512K ctx Apache 2.012.8K 34Aug 2026- Same category
- Similar size (2.5B vs 4B)
- Commercial use OK
LFM2.5 is a new family of hybrid models designed for on-device deployment. It builds on the LFM2 architecture with extended pre-training and reinforcement learning.
2.7B125K ctx lfm1.07.4K 53Aug 2026- Same category
- Similar size (2.7B vs 4B)