selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

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.

Compare lumi-4b with the top 3
the original
ceia-br logo

lumi-4b

ceia-br·Chat

≈4.2 GB

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.

4B256K ctx Unknown licence
3.4KOct 2026gguf
--vram
any
--gpus
--quant
  1. nvidia logo≈4.2 GB

    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.

    4B256K ctx nvidia-nemotron-open-model-license
    1.9MMar 2026gguf
    • Same category
    • Similar size (4B vs 4B)
    • Commercial use OK
    • Widely used
    vs
  2. Qwen logo≈4.0 GB

    We introduce the updated version of the Qwen3-4B non-thinking mode, named Qwen3-4B-Instruct-2507, featuring the following key enhancements:

    4B256K ctx Apache 2.0
    2.6MAug 2025gguf
    • Same category
    • Similar size (4B vs 4B)
    • Commercial use OK
    • Widely used
    vs
  3. Qwen logo

    Qwen3-4B

    Qwen·Chat

    ≈4.0 GB

    Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.

    4B40K ctx Apache 2.0
    3.2MApr 2025ollama
    • Same category
    • Similar size (4B vs 4B)
    • Commercial use OK
    • Widely used
    vs
  4. Nanbeige logo

    Nanbeige4.2-3B

    Nanbeige·Chat

    ≈3.7 GB

    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.

    4.2B256K ctx Apache 2.0
    96.9KJul 2026gguf
    • Same category
    • Similar size (4.2B vs 4B)
    • Commercial use OK
    vs
  5. ibm-granite logo

    granite-4.1-3b

    ibm-granite·Chat

    ≈3.1 GB

    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.

    3.4B128K ctx Apache 2.0
    456.9KApr 2026ollama
    • Same category
    • Similar size (3.4B vs 4B)
    • Commercial use OK
    vs
  6. XHToken logo

    Spark-X2.6-4B

    XHToken·Chat

    ≈4.1 GB

    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.

    4.1B1M ctx Apache 2.0
    29.1KOct 2026gguf
    • Same category
    • Similar size (4.1B vs 4B)
    • Commercial use OK
    vs
  7. empero-ai logo

    Qwen3.8-4B-Distill

    empero-ai·Chat

    ≈4.3 GB

    Qwen3.8-4B is a full-parameter distillation of Qwen3.8 2.4T A95B into the Qwen3.5-4B architecture. The student was trained on ~45,000 curated teacher traces from our internal Qwen3.8 distillation datasets — dense chain-of-thought spanning mathematics, general reasoning, and instruction following…

    4.7B256K ctx Apache 2.0
    130.8KAug 2026gguf
    • Same category
    • Similar size (4.7B vs 4B)
    • Commercial use OK
    vs
  8. NVIDIA-Nemotron-Labs-Dory1.0-4B-BF16 is a looped (latent-recurrent) large language model (LLM) trained by NVIDIA. The model employs a normalized-GPT architecture with interleaved hybrid sliding window attention (SWA) and full attention layers.

    4B256K ctx openmdw-1.1
    15KOct 2026gguf
    • Same category
    • Similar size (4B vs 4B)
    • Commercial use OK
    vs
  9. ibm-granite logo

    granite-4.2-3b

    ibm-granite·Chat

    ≈3.3 GB

    Granite is a family of open-source large language models developed by IBM, designed for enterprise and research use. Granite models are built to be versatile, safe, and efficient — covering a range of sizes and capabilities from compact edge-deployable models to large-scale reasoning systems.

    3.7B128K ctx Apache 2.0
    42.4KAug 2026ollama
    • Same category
    • Similar size (3.7B vs 4B)
    • Commercial use OK
    vs
  10. openbmb logo

    MiniCPM5-2B

    openbmb·Chat

    ≈2.3 GB

    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.

    2.5B128K ctx Apache 2.0
    1MSept 2026gguf
    • Same category
    • Similar size (2.5B vs 4B)
    • Commercial use OK
    vs
  11. nvidia logo≈3.6 GB

    Nemotron-Labs-Diffusion is a tri-mode language model that supports both AR decoding and diffusion-based parallel decoding by simply switching the attention pattern of the same model during inference.

    3.8B256K ctx nvidia-nemotron-open-model-license
    59.4KMar 2026gguf
    • Same category
    • Similar size (3.8B vs 4B)
    • Commercial use OK
    vs
  12. Qwen logo

    Qwen3-4B-Base

    Qwen·Chat

    ≈4.0 GB

    Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.

    4B32K ctx Apache 2.0
    435.4KApr 2025gguf
    • Same category
    • Similar size (4B vs 4B)
    • Commercial use OK
    vs
  13. XHToken logo

    Spark-X2.5-4B

    XHToken·Chat

    ≈4.1 GB

    We are introducing Spark-X2.5-4B and Spark-X2.5-1.7B, two compact, general-purpose language models designed to make capable AI more practical, efficient, and accessible.

    4.1B1M ctx Apache 2.0
    8.7KAug 2026gguf
    • Same category
    • Similar size (4.1B vs 4B)
    • Commercial use OK
    vs
  14. NVIDIA-Nemotron-Labs-Dory1.0-4B-MXFP8 is a looped (latent-recurrent) large language model (LLM) trained by NVIDIA. The model employs a normalized-GPT architecture with interleaved hybrid sliding window attention (SWA) and full attention layers.

    4B256K ctx openmdw-1.1
    2.5KOct 2026
    • Same category
    • Similar size (4B vs 4B)
    • Commercial use OK
    vs
  15. microsoft logo

    Phi-4-mini-instruct

    microsoft·Chat

    ≈3.8 GB

    Phi-4-mini-instruct is a lightweight open model built upon synthetic data and filtered publicly available websites - with a focus on high-quality, reasoning dense data. The model belongs to the Phi-4 model family and supports 128K token context length.

    3.8B128K ctx MIT
    332.5KFeb 2025ollama
    • Same category
    • Similar size (3.8B vs 4B)
    • Commercial use OK
    vs
  16. XHToken logo

    Spark-X2.6-4B-Base

    XHToken·Chat

    ≈4.1 GB

    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.

    4.1B1M ctx Apache 2.0
    2.4KOct 2026gguf
    • Same category
    • Similar size (4.1B vs 4B)
    • Commercial use OK
    vs
  17. Nanbeige logo

    Nanbeige4.1-3B

    Nanbeige·Chat

    ≈3.3 GB

    Nanbeige4.1-3B is built upon Nanbeige4-3B-Base and represents an enhanced iteration of our previous reasoning model, Nanbeige4-3B-Thinking-2511, achieved through further post-training optimization with supervised fine-tuning (SFT) and reinforcement learning (RL).

    3.9B256K ctx Apache 2.0
    9.3KFeb 2026gguf
    • Same category
    • Similar size (3.9B vs 4B)
    • Commercial use OK
    vs
  18. TokenRhythm logo

    NeoHorse-1-4B

    TokenRhythm·Chat

    ≈4.0 GB

    NeoHorse-1-4B is a 4B causal language model and an initial prototype on the path toward recursive self-improvement (RSI). It is post-trained from Qwen3.5-4B for text-based agent harnesses, tool use, coding, and instruction following.

    4.2B256K ctx Apache 2.0
    3KSept 2026gguf
    • Same category
    • Similar size (4.2B vs 4B)
    • Commercial use OK
    vs
  19. IFM logo≈4.6 GB

    K2-Horizon-3.7B is the small dense member of the K2-Horizon family: a 3.7B-core decoder-only model with a 512K context window.

    5.1B512K ctx Apache 2.0
    15.5KSept 2026gguf
    • Same category
    • Similar size (5.1B vs 4B)
    • Commercial use OK
    vs
  20. CohereLabs logo

    tiny-aya-global

    CohereLabs·Chat

    ≈3.1 GB

    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 4.0
    26.5KFeb 2026gguf
    • Same category
    • Similar size (3.3B vs 4B)
    vs
  21. microsoft logo

    UserRepGen-4B

    microsoft·Chat

    ≈4.2 GB

    UserRepGen-4B is a fine-tuned Qwen3.5-4B language model that generates a structured, natural-language user representation from a multi-turn user-assistant conversation.

    4.5B256K ctx MIT
    2.7KOct 2026gguf
    • Same category
    • Similar size (4.5B vs 4B)
    • Commercial use OK
    vs
  22. HuggingFaceTB logo

    SmolLM3-3B

    HuggingFaceTB·Chat

    ≈2.9 GB

    SmolLM3 is a 3B parameter language model designed to push the boundaries of small models. It supports dual mode reasoning, 6 languages and long context. SmolLM3 is a fully open model that offers strong performance at the 3B–4B scale.

    3.1B64K ctx Apache 2.0
    260.8KJul 2025gguf
    • Same category
    • Similar size (3.1B vs 4B)
    • Commercial use OK
    vs
  23. LiquidAI logo

    LFM2.5-2.6B

    LiquidAI·Chat

    ≈2.6 GB

    LFM2.5-2.6B is part of LFM2.5, a family of hybrid models designed for on-device deployment. It builds on the LFM2 architecture with a 128K context window and agentic post-training.

    2.7B128K ctx lfm1.0
    44.8KJul 2026gguf
    • Same category
    • Similar size (2.7B vs 4B)
    vs
  24. ibm-granite logo≈3.3 GB

    Model Summary: Granite-4.1-3B-LLMON-preview is a 3B parameter instruct model fine-tuned from Granite-4.1-3B-base to consume prompts in LLMON (LLM Object Notation) instead of a conventional chat template.

    3.7B128K ctx Apache 2.0
    1.8KOct 2026gguf
    • Same category
    • Similar size (3.7B vs 4B)
    • Commercial use OK
    vs