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 ZwZ2-27B

120 open-weight models that do the same job as ZwZ2-27B (27.4B vision), 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
inclusionAI logo

ZwZ2-27B

inclusionAI·Vision

≈19 GB

ZwZ2-27B is the Qwen3.6-27B-based model in Zooming-without-Zooming-v2 (ZwZ2). It learns to connect visual evidence across regions through comparison, aggregation, matching, and multi-step reasoning. Training uses full images; inference requires no zooming tools.

27.4B256K ctx Unknown licence
1.5KOct 2026gguf
--vram
any
--gpus
--quant
  1. tencent logo

    Hy-Embodied-VLM-1.0

    tencent·Vision

    ≈19 GB

    Building capable embodied agents requires not only multimodal perception and understanding, but also agentic capabilities for reasoning about actions, adapting to evolving situations, and interacting with the physical world.

    30.5B32K ctx Apache 2.0
    544Jul 2026
    • Same category
    • Similar size (30.5B vs 27.4B)
    • Commercial use OK
    vs
  2. bottlecapai logo

    ThinkingCap-Qwen3.6-27B

    bottlecapai·Vision

    ≈19 GB

    Capability of Qwen3.6-27B with 50% less thinking tokens on average, and over 90% less in best cases. Achieved via finetuning Qwen3.6-27B (Qwen Team, 2026) with state-of-the-art algorithms on a curated set of problems of various domains and difficulty.

    27.4B256K ctx Apache 2.0
    179Jul 2026gguf
    • Same category
    • Similar size (27.4B vs 27.4B)
    • Commercial use OK
    vs
  3. Qwen logo

    Qwen3.5-9B

    Qwen·Vision

    ≈7.2 GB

    Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.

    9.7B256K ctx Apache 2.0
    6.2MFeb 2026ollama
    • Same category
    • Smaller (9.7B vs 27.4B)
    • Commercial use OK
    • Widely used
    vs
  4. google logo

    gemma-3-27b-pt

    google·Vision

    ≈21 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.

    27.4BGemma Terms of Use
    6.3KMar 2025gguf
    • Same category
    • Similar size (27.4B vs 27.4B)
    vs
  5. Agnes-AI logo

    Agnes-3.0-Flash

    Agnes-AI·Vision

    ≈22 GB

    This repository contains an earlier open-weight Preview checkpoint of Agnes 3.0 Flash. It is distinct from the newer production/API checkpoint listed on Artificial Analysis. The Preview release has 33B parameters and a context window of 262,144 tokens.

    33.1B256K ctx Apache 2.0
    447Sept 2026gguf
    • Same category
    • Similar size (33.1B vs 27.4B)
    • Commercial use OK
    vs
  6. baidu logo≈19 GB

    Here is an example of how to use the transformers library for inference:

    29.7BA3B128K ctx Apache 2.0
    1.4KNov 2025
    • Same category
    • Similar size (29.7B vs 27.4B)
    • Commercial use OK
    vs
  7. internlm logo

    Intern-S2-Mobius

    internlm·Vision

    ≈22 GB

    We introduce Intern-S2-Mobius, a 35B foundation model built on the Mobius-v0 architecture realized by Xtuner and LMDeploy.

    36B256K ctx Apache 2.0
    1KJul 2026
    • Same category
    • Similar size (36B vs 27.4B)
    • Commercial use OK
    vs
  8. nvidia logo≈22 GB

    NVIDIA-Ising-Calibration-1 analyzes quantum computing calibration experiment plots and generates structured technical text across six analysis question categories. NVIDIA-Ising-Calibration-1 was developed by NVIDIA as a quantum calibration vision-language model built on Qwen3.5-35B-A3B.

    35BA3B256K ctx NVIDIA Open Model License
    2KMar 2026gguf
    • Same category
    • Similar size (35B vs 27.4B)
    • Commercial use OK
    vs
  9. internlm logo

    Intern-S2-Preview

    internlm·Vision

    ≈23 GB

    We introduce Intern-S2-Preview, an efficient 35B scientific multimodal foundation model. Beyond conventional parameter and data scaling, Intern-S2-Preview explores task scaling: increasing the difficulty, diversity, and coverage of scientific tasks to further unlock model capabilities.

    36.1BA4.9B256K ctx Apache 2.0
    1.6KMay 2026
    • Same category
    • Similar size (36.1B vs 27.4B)
    • Commercial use OK
    vs
  10. OpenGVLab logo≈20 GB

    We introduce InternVL3.5, a new family of open-source multimodal models that significantly advances versatility, reasoning capability, and inference efficiency along the InternVL series.

    30.8BA3B40K ctx Apache 2.0
    2.4KAug 2025
    • Same category
    • Similar size (30.8B vs 27.4B)
    • Commercial use OK
    vs
  11. rhymes-ai logo

    Aria

    rhymes-ai·Vision

    ≈18 GB

    [Dec 1, 2024] We have released the base models (with native multimodal pre-training) for Aria (Aria-Base-8K and Aria-Base-64K) for research purposes and continue training.

    25.3BA4.6B64K ctx Apache 2.0
    19.2KSept 2024
    • Same category
    • Similar size (25.3B vs 27.4B)
    • Commercial use OK
    vs
  12. tencent logo

    WeMM-Video-35BA3B

    tencent·Vision

    ≈23 GB

    WeMM-Video-35B-A3B is the mixture-of-experts (35B total, 3B active) checkpoint in the WeMM-Video family. WeMM-Video is a Chinese–English bilingual family for joint audio-visual understanding and reasoning. It accepts text, images, audio, and video and generates text responses.

    36.2BA5B256K ctx Apache 2.0
    239Nov 2026gguf
    • Same category
    • Similar size (36.2B vs 27.4B)
    • Commercial use OK
    • Newer release
    vs
  13. inclusionAI logo

    UI-Venus-1.5-30B-A3B

    inclusionAI·Vision

    ≈20 GB

    This repository contains the UI-Venus model from the report UI-Venus-1.5 Technical Report. UI-Venus 1.5 is a unified, end-to-end GUI Agent designed for robust real-world applications. The model family includes two dense (2B/8B) and one MoE (30B-A3B) variants to meet various downstream scenarios.

    31.1BA3B256K ctx Apache 2.0
    339Feb 2026gguf
    • Same category
    • Similar size (31.1B vs 27.4B)
    • Commercial use OK
    vs
  14. deepseek-ai logo

    deepseek-vl2

    deepseek-ai·Vision

    ≈19 GB

    Introducing DeepSeek-VL2, an advanced series of large Mixture-of-Experts (MoE) Vision-Language Models that significantly improves upon its predecessor, DeepSeek-VL.

    27.5BDeepSeek License
    2.3KDec 2024
    • Same category
    • Similar size (27.5B vs 27.4B)
    vs
  15. Qwen logo≈6.8 GB

    Meet Qwen3-VL — the most powerful vision-language model in the Qwen series to date.

    8.8B256K ctx Apache 2.0
    6.1MOct 2025ollama
    • Same category
    • Smaller (8.8B vs 27.4B)
    • Commercial use OK
    • Widely used
    vs
  16. baidu logo≈18 GB

    The advanced capabilities of the ERNIE 4.5 models, particularly the MoE-based A47B and A3B series, are underpinned by several key technical innovations:

    29.4BA3B128K ctx Apache 2.0
    720Jun 2025
    • Same category
    • Similar size (29.4B vs 27.4B)
    • Commercial use OK
    vs
  17. OpenGVLab logo

    InternVL3_5-38B

    OpenGVLab·Vision

    ≈25 GB

    We introduce InternVL3.5, a new family of open-source multimodal models that significantly advances versatility, reasoning capability, and inference efficiency along the InternVL series.

    38.4B40K ctx Apache 2.0
    4.3KAug 2025gguf
    • Same category
    • Similar size (38.4B vs 27.4B)
    • Commercial use OK
    vs
  18. Zyphra logo

    ZAYA1-VL-8B

    Zyphra·Vision

    ≈6.6 GB

    ZAYA1-VL-8B is a vision-language model (VLM) built upon Zyphra's ZAYA1-8B LLM. It has state-of-the-art performance among VLMs for its size and inference efficiency.

    9.7B32K ctx Apache 2.0
    339.4KApr 2026gguf
    • Same category
    • Smaller (9.7B vs 27.4B)
    • Commercial use OK
    vs
    • Same category
    • Similar size (40B vs 27.4B)
    • Commercial use OK
    vs
  19. microsoft logo≈11 GB

    Developer: Microsoft Corporation Authorized Representative: Microsoft Ireland Operations Limited, 70 Sir John Rogerson's Quay, Dublin 2, D02 R296, Ireland Release Date: March 4, 2026 License: MIT Parameters: 15B Context Length: 16,384 tokens Inputs: Text and Images Outputs: Text Training GPUs: 240…

    15.1B32K ctx MIT
    9.7KJan 2026gguf
    • Same category
    • Similar size (15.1B vs 27.4B)
    • Commercial use OK
    vs
  20. Skywork logo

    Skywork-R1V-38B

    Skywork·Vision

    ≈25 GB

    Skywork-R1V-38B is an open-weight vision model with 38.4B parameters from Skywork, published on Hugging Face.

    38.4B128K ctx MIT
    7.8KMar 2025
    • Same category
    • Similar size (38.4B vs 27.4B)
    • Commercial use OK
    vs
  21. moonshotai logo

    Kimi-VL-A3B-Instruct

    moonshotai·Vision

    ≈13 GB

    We present Kimi-VL, an efficient open-source Mixture-of-Experts (MoE) vision-language model (VLM) that offers advanced multimodal reasoning, long-context understanding, and strong agent capabilities—all while activating only 2.8B parameters in its language decoder (Kimi-VL-A3B).

    16.4BA3B128K ctx MIT
    28.5KApr 2025gguf
    • Same category
    • Similar size (16.4B vs 27.4B)
    • Commercial use OK
    vs
  22. omlab logo

    VLX-Seek-1.5-10B

    omlab·Vision

    ≈7.4 GB

    VLX-Seek-1.5-10B is the open-source 10B model in the VLX-Seek 1.5 family, designed for fine-grained perception and visual grounding in embodied scenarios.

    10B256K ctx Apache 2.0
    73.4KJul 2026gguf
    • Same category
    • Smaller (10B vs 27.4B)
    • Commercial use OK
    vs
  23. baidu logo≈18 GB

    The advanced capabilities of the ERNIE 4.5 models, particularly the MoE-based A47B and A3B series, are underpinned by several key technical innovations:

    29.4BA3BApache 2.0
    257Jun 2025
    • Same category
    • Similar size (29.4B vs 27.4B)
    • Commercial use OK
    vs