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.

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
3.6KOct 2026gguf
--vram
any
--gpus
--quant
  1. A compact 9B reasoning model distilled from DeepSeek-V4-Pro in Max Effect mode

    9.4B256K ctx Apache 2.0
    24.5KOct 2026gguf
    • Same category
    • Larger (9.4B vs 4B)
    • Commercial use OK
    • Newer release
    vs
  2. ibm-granite logo

    granite-3.3-2b-instruct

    ibm-granite·Chat

    ≈2.6 GB

    Model Summary: Granite-3.3-2B-Instruct is a 2-billion parameter 128K context length language model fine-tuned for improved reasoning and instruction-following capabilities.

    2.5B128K ctx Apache 2.0
    41.1KApr 2025ollama
    • Same category
    • Similar size (2.5B vs 4B)
    • Commercial use OK
    vs
  3. nvidia logo≈6.6 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.

    8.5B256K ctx nvidia-nemotron-open-model-license
    49.7KMar 2026gguf
    • Same category
    • Larger (8.5B vs 4B)
    • Commercial use OK
    vs
  4. ServiceNow-AI logo

    Apriel-5B-Instruct

    ServiceNow-AI·Chat

    ≈4.2 GB

    Apriel is a family of models built for versatility, offering high throughput and efficiency across a wide range of tasks.

    4.8B16K ctx MIT
    3.9KApr 2025
    • Same category
    • Similar size (4.8B vs 4B)
    • Commercial use OK
    vs
  5. deepcogito logo≈3.5 GB

    The Cogito LLMs are instruction tuned generative models (text in/text out). All models are released under an open license for commercial use.

    3.6B128K ctx Llama 3.2 Community License
    1.6KMar 2025gguf
    • Same category
    • Similar size (3.6B vs 4B)
    vs
  6. ibm-granite logo

    granite-4.0-h-tiny

    ibm-granite·Chat

    ≈5.2 GB

    📣 Update [10-07-2025]: Added a default system prompt to the chat template to guide the model towards more professional, accurate, and safe responses.

    6.9BA1.5B128K ctx Apache 2.0
    27.4KSept 2025gguf
    • Same category
    • Similar size (6.9B vs 4B)
    • Commercial use OK
    vs
  7. Qwen logo≈6.5 GB

    Qwen3Guard is a series of safety moderation models built upon Qwen3 and trained on a dataset of 1.19 million prompts and responses labeled for safety.

    8.2B32K ctx Apache 2.0
    110.5KSept 2025gguf
    • Same category
    • Larger (8.2B vs 4B)
    • Commercial use OK
    vs
  8. google logo

    gemma-2-2b-it

    google·Chat

    ≈2.9 GB

    Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models.

    2.6B8K ctx Gemma Terms of Use
    170.2KJul 2024ollama
    • Same category
    • Similar size (2.6B vs 4B)
    vs
  9. speakleash logo≈4.1 GB

    Bielik-4.5B-v3-Instruct is a generative text model featuring 4.6 billion parameters. It is an instruct fine-tuned version of the Bielik-4.5B-v3.

    4.8BApache 2.0
    2.4KApr 2025gguf
    • Same category
    • Similar size (4.8B vs 4B)
    • Commercial use OK
    vs
  10. LiquidAI logo

    LFM2-2.6B

    LiquidAI·Chat

    ≈2.5 GB

    LFM2 is a new generation of hybrid models developed by Liquid AI, specifically designed for edge AI and on-device deployment. It sets a new standard in terms of quality, speed, and memory efficiency.

    2.6B125K ctx lfm1.0
    6.7KSept 2025gguf
    • Same category
    • Similar size (2.6B vs 4B)
    vs
  11. LiquidAI logo

    LFM2.5-8B-A1B

    LiquidAI·Chat

    ≈5.9 GB

    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.

    8.5BA1B125K ctx lfm1.0
    19.7KMay 2026ollama
    • Same category
    • Larger (8.5B vs 4B)
    vs
  12. tiiuae logo≈3.1 GB

    Falcon3 family of Open Foundation Models is a set of pretrained and instruct LLMs ranging from 1B to 10B parameters.

    3.2B32K ctx Falcon LLM License
    7.3KDec 2024ollama
    • Same category
    • Similar size (3.2B vs 4B)
    vs
  13. LiquidAI logo

    LFM3-1.7B-Base

    LiquidAI·Chat

    ≈1.9 GB

    LFM3-1.7B-Base is the pre-trained checkpoint behind LFM3-1.7B, a text model for local agents and specialized applications based on our new tri-hybrid architecture.

    1.7B256K ctx lfm1.0
    11.6KOct 2026ollama
    • Same category
    • Smaller (1.7B vs 4B)
    vs
  14. allenai logo

    Olmo-Hybrid-7B

    allenai·Chat

    ≈8.7 GB

    We expand on our Olmo model series by introducing Olmo Hybrid, a new 7B hybrid RNN model in the Olmo family. Olmo Hybrid dramatically outperforms Olmo 3 in final performance, consistently showing roughly 2x data efficiency on core evals over the course of our pretraining run.

    7.4B64K ctx Apache 2.0
    11.1KJan 2026
    • Same category
    • Similar size (7.4B vs 4B)
    • Commercial use OK
    vs
  15. nvidia logo

    Minitron-4B-Base

    nvidia·Chat

    ≈3.4 GB

    Minitron-4B-Base is a large language model (LLM) obtained by pruning Nemotron-4 15B; specifically, we prune model embedding size, number of attention heads, and MLP intermediate dimension.

    4B4K ctx NVIDIA Open Model License
    2.9KJul 2024gguf
    • Same category
    • Similar size (4B vs 4B)
    • Commercial use OK
    vs
  16. Qwen logo

    Qwen3-1.7B-Base

    Qwen·Chat

    ≈2.4 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.

    1.7B32K ctx Apache 2.0
    617.6KApr 2025gguf
    • Same category
    • Smaller (1.7B vs 4B)
    • Commercial use OK
    vs
  17. ibm-granite logo

    granite-4.0-h-micro-base

    ibm-granite·Chat

    ≈3.0 GB

    Model Summary: Granite-4.0-H-Micro-Base is a decoder-only, long-context language model designed for a wide range of text-to-text generation tasks. It also supports Fill-in-the-Middle (FIM) code completion through the use of specialized prefix and suffix tokens.

    3.2B128K ctx Apache 2.0
    708Sept 2025gguf
    • Same category
    • Similar size (3.2B vs 4B)
    • Commercial use OK
    vs
  18. LiquidAI logo

    LFM2-2.6B-Exp

    LiquidAI·Chat

    ≈2.5 GB

    LFM2-2.6B-Exp is an experimental checkpoint built on LFM2-2.6B using pure reinforcement learning.

    2.6B125K ctx lfm1.0
    2.3KDec 2025gguf
    • Same category
    • Similar size (2.6B vs 4B)
    vs
  19. arcee-ai logo

    Trinity-Nano-Preview

    arcee-ai·Chat

    ≈4.6 GB

    Trinity Nano Preview is a preview of Arcee AI's 6B MoE model with 1B active parameters. It is the small-sized model in our new Trinity family, a series of open-weight models for enterprise and tinkerers alike.

    6.1BA1B128K ctx openmdw-1.1
    2.7KDec 2025gguf
    • Same category
    • Similar size (6.1B vs 4B)
    • Commercial use OK
    vs
  20. ibm-granite logo

    granite-3.1-3b-a800m-base

    ibm-granite·Chat

    ≈3.0 GB

    Model Summary: Granite-3.1-3B-A800M-Base extends the context length of Granite-3.0-3B-A800M-Base from 4K to 128K using a progressive training strategy by increasing the supported context length in increments while adjusting RoPE theta until the model has successfully adapted to desired length of…

    3.3BA800M128K ctx Apache 2.0
    4.5KDec 2024gguf
    • Same category
    • Similar size (3.3B vs 4B)
    • Commercial use OK
    vs
  21. LGAI-EXAONE logo

    EXAONE-3.5-7.8B-Instruct

    LGAI-EXAONE·Chat

    ≈6.1 GB

    We introduce EXAONE 3.5, a collection of instruction-tuned bilingual (English and Korean) generative models ranging from 2.4B to 32B parameters, developed and released by LG AI Research.

    7.8B32K ctx exaone
    395.6KDec 2024ollama
    • Same category
    • Similar size (7.8B vs 4B)
    vs
  22. llm-jp logo≈6.6 GB

    LLM-jp-4 is a series of large language models developed by the Research and Development Center for Large Language Models at the National Institute of Informatics.

    8.6B64K ctx Apache 2.0
    22.4KMar 2026gguf
    • Same category
    • Larger (8.6B vs 4B)
    • Commercial use OK
    vs
  23. Qwen logo

    Qwen2.5-7B

    Qwen·Chat

    ≈5.5 GB

    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:

    7.6B128K ctx Apache 2.0
    531.9KSept 2024gguf
    • Same category
    • Similar size (7.6B vs 4B)
    • Commercial use OK
    vs
  24. zai-org logo

    glm-edge-4b-chat

    zai-org·Chat

    ≈4.0 GB

    The usage of this model’s weights is subject to the terms outlined in the LICENSE.

    4.3B8K ctx GLM-4 License
    2KNov 2024gguf
    • Same category
    • Similar size (4.3B vs 4B)
    vs