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 Hy4-flash

120 open-weight models that do the same job as Hy4-flash (220B 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
tencent logo

Hy4-flash

tencent·Vision

≈134 GB

Hy4 flash is our most cost-effective Hy model to date, featuring 220B total parameters (including 64B Engram parameters) with just 12B active parameters per token. It natively accepts text and image inputs, generates text, and supports a context window of up to 1M tokens.

220B1M ctx tencent-hy-license
5.6KNov 2026gguf
--vram
any
--gpus
--quant
  1. moonshotai logo

    Kimi-K2.5

    moonshotai·Vision

    ≈632 GB

    Kimi K2.5 is an open-source, native multimodal agentic model built through continual pretraining on approximately 15 trillion mixed visual and text tokens atop Kimi-K2-Base.

    1.03TA33.3B256K ctx modified-mit
    389.7KJan 2026gguf
    • Same category
    • Larger (1.03T vs 220B)
    • Commercial use OK
    vs
  2. baidu logo≈253 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:

    424BA47B128K ctx Apache 2.0
    428Jun 2025
    • Same category
    • Similar size (424B vs 220B)
    • Commercial use OK
    vs
  3. Cactus-Compute logo

    mirage

    Cactus-Compute·Vision

    ? GB

    The package downloads and caches the engine and weights on first use. This repository provides mirage.cact, mirage.safetensors and the model config.

    ?BApache 2.0
    56.3KNov 2026
    • Same category
    • Commercial use OK
    vs
  4. Qwen logo

    Qwen3.8-27B

    Qwen·Vision

    ≈19 GB

    Following the widespread community adoption of the Qwen3.5 and Qwen3.6 series, we are pleased to introduce Qwen3.8, the most capable generation in the Qwen open-model family to date.

    27.8B256K ctx Apache 2.0
    6.3MAug 2026ollama
    • Same category
    • Smaller (27.8B vs 220B)
    • Commercial use OK
    • Widely used
    vs
  5. google logo

    gemma-4-31B-it

    google·Vision

    ≈27 GB

    Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants.

    31.3B256K ctx Apache 2.0
    6MMar 2026ollama
    • Same category
    • Smaller (31.3B vs 220B)
    • Commercial use OK
    • Widely used
    vs
  6. mistralai logo

    Mistral-Large-4-1T-A52B

    mistralai·Vision

    ≈602 GB

    Mistral Large 4 is our largest and most capable model to date: a 1 trillion-parameter natively multimodal Mixture-of-Experts model with 49 billion active parameters per token (52 billion including embeddings and output layers), trained from scratch on 3,800 NVIDIA Grace Blackwell GPUs in Mistral's…

    1TA52B1M ctx Custom licence
    20.6KSept 2026gguf
    • Same category
    • Larger (1T vs 220B)
    vs
  7. The advanced capabilities of the ERNIE 4.5 models, particularly the MoE-based A47B and A3B series, are underpinned by several key technical innovations:

    424BA47B128K ctx Apache 2.0
    228Jun 2025
    • Same category
    • Similar size (424B vs 220B)
    • Commercial use OK
    vs
  8. baidu logo≈253 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:

    424BA47B128K ctx Apache 2.0
    245Jun 2025
    • Same category
    • Similar size (424B vs 220B)
    • Commercial use OK
    vs
  9. Qwen logo≈47 GB

    In the past five months since Qwen2-VL’s release, numerous developers have built new models on the Qwen2-VL vision-language models, providing us with valuable feedback. During this period, we focused on building more useful vision-language models.

    73.4B125K ctx Qwen License
    38.3KJan 2025ollama
    • Same category
    • Smaller (73.4B vs 220B)
    vs
  10. Qwen logo

    Qwen3.5-35B-A3B

    Qwen·Vision

    ≈22 GB

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

    36BA3B256K ctx Apache 2.0
    1.3MFeb 2026ollama
    • Same category
    • Smaller (36B vs 220B)
    • Commercial use OK
    vs
  11. OpenGVLab logo

    InternVL3-78B

    OpenGVLab·Vision

    ≈49 GB

    We introduce InternVL3, an advanced multimodal large language model (MLLM) series that demonstrates superior overall performance.

    78.4B32K ctx Qwen License
    11.2KApr 2025gguf
    • Same category
    • Smaller (78.4B vs 220B)
    vs
  12. meta-llama logo

    Llama-4-Maverick-17B-128E

    meta-llama·Vision

    ≈240 GB

    The Llama 4 collection of models are natively multimodal AI models that enable text and multimodal experiences. These models leverage a mixture-of-experts architecture to offer industry-leading performance in text and image understanding.

    402BA17B256K ctx Llama 4 Community License
    206Apr 2025gguf
    • Same category
    • Similar size (402B vs 220B)
    vs
  13. tencent logo

    Hy4

    tencent·Vision

    ≈461 GB

    Hy4 is a new-generation Mixture-of-Experts (MoE) flagship model developed by the Tencent Hy Team, supporting text, image, and video understanding. The language backbone comprises 770B total parameters, of which 49B are activated per token.

    774BA53.6B1M ctx tencent-hy-license
    974Oct 2026gguf
    • Same category
    • Larger (774B vs 220B)
    vs
  14. tencent logo

    Hy4-base

    tencent·Vision

    ≈459 GB

    Hy4 is a new-generation Mixture-of-Experts (MoE) flagship model developed by the Tencent Hy Team, supporting text, image, and video understanding. The language backbone comprises 770B total parameters, of which 49B are activated per token.

    771BA50.3B1M ctx tencent-hy-license
    482Nov 2026
    • Same category
    • Larger (771B vs 220B)
    • Newer release
    vs
  15. PaddlePaddle logo

    UVDoc

    PaddlePaddle·Vision

    ? GB

    The main purpose of text image correction is to carry out geometric transformation on the image to correct the document distortion, inclination, perspective deformation and other problems in the image, so that the subsequent text recognition can be more accurate.

    ?BApache 2.0
    450.9KJun 2025
    • Same category
    • Commercial use OK
    vs
  16. Qwen logo

    Qwen3.6-27B

    Qwen·Vision

    ≈19 GB

    Following the February release of the Qwen3.5 series, we're pleased to share the first open-weight variant of Qwen3.6. Built on direct feedback from the community, Qwen3.6 prioritizes stability and real-world utility, offering developers a more intuitive, responsive, and genuinely productive coding…

    27.8B256K ctx Apache 2.0
    2.4MApr 2026ollama
    • Same category
    • Smaller (27.8B vs 220B)
    • Commercial use OK
    • Widely used
    vs
  17. internlm logo

    Intern-S1-Pro

    internlm·Vision

    ≈546 GB

    💻Github Repo • 🤗Model Collections • 📜Technical Report • 💬Online Chat

    917BA22.9B256K ctx Apache 2.0
    11.6KFeb 2026
    • Same category
    • Larger (917B vs 220B)
    • Commercial use OK
    vs
  18. PaddlePaddle logo

    PP-LCNet_x1_0_doc_ori

    PaddlePaddle·Vision

    ? GB

    The Document Image Orientation Classification Module is primarily designed to distinguish the orientation of document images and correct them through post-processing.

    ?BApache 2.0
    326.3KJun 2025
    • Same category
    • Commercial use OK
    vs
  19. google logo

    gemma-4-26B-A4B-it

    google·Vision

    ≈18 GB

    Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on E2B, E4B, and 12B) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants.

    25.8BA4B256K ctx Apache 2.0
    4.8MMar 2026ollama
    • Same category
    • Smaller (25.8B vs 220B)
    • Commercial use OK
    • Widely used
    vs
  20. nvidia logo

    NVLM-D-72B

    nvidia·Vision

    ≈50 GB

    This family of models performs vision-language and text-only tasks including optical character recognition, multimodal reasoning, localization, common sense reasoning, world knowledge utilization, and coding.

    79.4B32K ctx CC BY-NC 4.0
    7.9KSept 2024
    • Same category
    • Smaller (79.4B vs 220B)
    vs
  21. Qwen logo≈47 GB

    We're excited to unveil Qwen2-VL, the latest iteration of our Qwen-VL model, representing nearly a year of innovation.

    73.4B32K ctx Qwen License
    14.9KSept 2024gguf
    • Same category
    • Smaller (73.4B vs 220B)
    vs
  22. Qwen logo

    Qwen3.5-27B

    Qwen·Vision

    ≈19 GB

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

    27.8B256K ctx Apache 2.0
    1.6MFeb 2026ollama
    • Same category
    • Smaller (27.8B vs 220B)
    • Commercial use OK
    vs
  23. ByteDance-Seed logo

    UI-TARS-72B-DPO

    ByteDance-Seed·Vision

    ≈47 GB

    UI-TARS is a next-generation native GUI agent model designed to interact seamlessly with graphical user interfaces (GUIs) using human-like perception, reasoning, and action capabilities.

    73.4B32K ctx Apache 2.0
    2.9KJan 2025gguf
    • Same category
    • Smaller (73.4B vs 220B)
    • Commercial use OK
    vs
  24. meta-models logo

    Muse-Glimmer-30B

    meta-models·Vision

    ≈19 GB

    Muse Glimmer is a 30-billion-parameter causal language model with a dedicated perception encoder, distilled from Muse Spark and purpose-built for autonomous agentic tasks on consumer hardware.

    29.8B128K ctx Apache 2.0
    149.3KAug 2026ollama
    • Same category
    • Smaller (29.8B vs 220B)
    • Commercial use OK
    vs