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-base

120 open-weight models that do the same job as Hy4-base (771B 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-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
68Nov 2026
--vram
any
--gpus
--quant
  1. Qwen logo

    Qwen3.5-4B-Base

    Qwen·Vision

    ≈4.3 GB

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

    4.7B256K ctx Apache 2.0
    504KFeb 2026gguf
    • Same category
    • Smaller (4.7B vs 771B)
    • Commercial use OK
    vs
  2. 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
    337.9KApr 2026gguf
    • Same category
    • Smaller (9.7B vs 771B)
    • Commercial use OK
    vs
  3. meta-models logo

    Muse-Glimmer-30B-assistant

    meta-models·Vision

    ≈2.2 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.

    2.6B128K ctx Apache 2.0
    129.6KAug 2026
    • Same category
    • Smaller (2.6B vs 771B)
    • Commercial use OK
    vs
  4. Qwen logo≈2.6 GB

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

    2.1B256K ctx Apache 2.0
    1.1MOct 2025ollama
    • Same category
    • Smaller (2.1B vs 771B)
    • Commercial use OK
    vs
  5. Cloudflare logo

    clef

    Cloudflare·Vision

    ≈19 GB

    Clef is a 27B multimodal model that turns a state and a schema of typed questions into decisions. It reads the state as text, JSON, images, or video, and returns a probability for every allowed option of every question in a single forward pass.

    27.4B256K ctx Apache 2.0
    78.2KSept 2026ollama
    • Same category
    • Smaller (27.4B vs 771B)
    • Commercial use OK
    vs
  6. interfaze-ai logo

    interfaze-1-lite

    interfaze-ai·Vision

    ? GB

    Interfaze 1.1 Lite is a mixture-of-architectures (MoA) model for developer workloads: reading documents, transcribing speech, locating objects and interface elements, and answering over all of it with structured output.

    ?Binterfaze-open
    137Oct 2026
    • Same category
    vs
  7. Qwen logo

    Qwen3.5-2B-Base

    Qwen·Vision

    ≈2.2 GB

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

    2.3B256K ctx Apache 2.0
    322.9KFeb 2026gguf
    • Same category
    • Smaller (2.3B vs 771B)
    • Commercial use OK
    vs
  8. nvidia logo

    Cosmos-Reason2-2B

    nvidia·Vision

    ≈2.5 GB

    NVIDIA Cosmos Reason 2 is an open, customizable, 2B-parameter reasoning vision language model (VLM) for physical AI and robotics that enables robots and vision AI agents to reason like humans, using prior knowledge, physics understanding and common sense to understand and act in the real world.

    2.4B256K ctx NVIDIA Open Model License
    491.3KDec 2025gguf
    • Same category
    • Smaller (2.4B vs 771B)
    • Commercial use OK
    vs
  9. 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
    88.2KJul 2026gguf
    • Same category
    • Smaller (10B vs 771B)
    • Commercial use OK
    vs
  10. openbmb logo

    MiniCPM-V-4.6

    openbmb·Vision

    ≈1.6 GB

    A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone

    1.3B256K ctx Apache 2.0
    189.7KApr 2026ollama
    • Same category
    • Smaller (1.3B vs 771B)
    • Commercial use OK
    vs
  11. Qwen logo≈5.8 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.

    8.3B125K ctx Apache 2.0
    5.3MJan 2025ollama
    • Same category
    • Smaller (8.3B vs 771B)
    • Commercial use OK
    • Widely used
    vs
  12. Qwen logo

    Qwen3.5-9B-Base

    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
    264.8KFeb 2026gguf
    • Same category
    • Smaller (9.7B vs 771B)
    • Commercial use OK
    vs
  13. black-forest-labs logo

    flux-3-image-klein-pup

    black-forest-labs·Vision

    ≈7.1 GB

    FLUX 3 Image Klein Prompt Upsampler is a 9B vision-language model that expands a short image request into the detailed prompt FLUX 3 Image Klein works best with. It takes the request, up to 10 reference images and an optional list of bounding boxes, and returns one prompt.

    9.4B256K ctx Apache 2.0
    39.6KOct 2026gguf
    • Same category
    • Smaller (9.4B vs 771B)
    • Commercial use OK
    vs
  14. tencent logo

    WeVisOne-VL-4B

    tencent·Vision

    ≈4.5 GB

    WeVisOne-VL (WeChat Vision All-in-One) is a unified vision-language model developed by WeChat Vision, Tencent Inc. Built upon Qwen3-VL, it unifies object-level perception (open-vocabulary object detection, referring localization, object counting, object pointing, and OCR) and pixel-level perception…

    4.8B256K ctx Apache 2.0
    35.2KOct 2026gguf
    • Same category
    • Smaller (4.8B vs 771B)
    • Commercial use OK
    vs
  15. meta-llama logo≈66 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.

    109BA18B10M ctx Llama 4 Community License
    109.8KApr 2025gguf
    • Same category
    • Smaller (109B vs 771B)
    vs
  16. Cloudflare logo

    clef-flash

    Cloudflare·Vision

    ≈7.1 GB

    Clef-Flash is a 9B multimodal model that turns a state and a schema of typed questions into decisions. It reads the state as text, JSON, images, or video, and returns a probability for every allowed option of every question in a single forward pass.

    9.4B256K ctx Apache 2.0
    33.4KSept 2026ollama
    • Same category
    • Smaller (9.4B vs 771B)
    • Commercial use OK
    vs
  17. LiquidAI logo

    LFM3-VL-2B

    LiquidAI·Vision

    ≈2.0 GB

    LFM3-VL-2B is a vision-language model for on-device visual understanding. It takes text, images, and video as input, and combines the LFM3 language model with a new in-house vision encoder.

    1.9B125K ctx lfm1.0
    28.7KOct 2026ollama
    • Same category
    • Smaller (1.9B vs 771B)
    vs
  18. XiaomiMiMo logo

    MiMo-V2.6-Distill-Qwen-9B

    XiaomiMiMo·Vision

    ≈7.1 GB

    MiMo-V2.6-Distill-Qwen-9B is a 9B agentic model developed by Xiaomi MiMo through supervised fine-tuning of Qwen3.5-9B on MiMo-generated data. It covers coding, general-purpose agent tasks, visual coding, and cybersecurity.

    9.4B256K ctx MIT
    31.5KSept 2026gguf
    • Same category
    • Smaller (9.4B vs 771B)
    • Commercial use OK
    vs
  19. google logo

    medgemma-1.5-4b-it

    google·Vision

    ≈4.1 GB

    * License: The use of MedGemma is governed by the Health AI Developer Foundations terms of use. MedGemma has not been evaluated or optimized for multi-turn applications.

    4.3B128K ctx health-ai-developer-foundations
    209.6KJan 2026ollama
    • Same category
    • Smaller (4.3B vs 771B)
    vs
  20. h2oai logo

    h2o-lightning-4b

    h2oai·Vision

    ≈4.2 GB

    H2O-Lightning-4B is a 4B-parameter decision model from H2O.ai, built on Qwen/Qwen3.5-4B. It answers typed decision questions about a record (a document, a ticket, a policy, a conversation) and about images that come with the record (photos, screenshots, scanned documents, charts).

    4.5B256K ctx Apache 2.0
    26.2KOct 2026gguf
    • Same category
    • Smaller (4.5B vs 771B)
    • Commercial use OK
    vs
  21. LiquidAI logo

    LFM2.5-VL-3B

    LiquidAI·Vision

    ≈2.8 GB

    LFM2.5-VL-3B is a multimodal variant of LFM2.5, a family of hybrid models designed for on-device deployment. It builds on LFM2-VL-3B with further mid- and post-training.

    3.1B32K ctx lfm1.0
    36.4KAug 2026gguf
    • Same category
    • Smaller (3.1B vs 771B)
    vs
  22. Qwen logo≈20 GB

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

    31.1BA3B256K ctx Apache 2.0
    432.4KSept 2025ollama
    • Same category
    • Smaller (31.1B vs 771B)
    • Commercial use OK
    vs
  23. LiquidAI logo

    d1-3B

    LiquidAI·Vision

    ≈2.8 GB

    d1-3B is a 3B parameter decision model built on LFM2.5-VL-3B. You give it a state (text, JSON, images, or a mix) and a set of questions. It returns calibrated, typed answers in one forward pass with zero output tokens.

    3.1B32K ctx lfm1.0
    20.3KOct 2026gguf
    • Same category
    • Smaller (3.1B vs 771B)
    vs
  24. nvidia logo

    Nemotron-3.5-Parse

    nvidia·Vision

    ≈1.2 GB

    Nemotron 3.5 Parse is a document image parser that produces structured, machine-readable text with semantic layout classes, bounding boxes, and reading order. It combines a C-RADIOv4-SO400M vision encoder with a 10-layer mBART decoder that consumes the visual sequence as a causal prefix.

    631M16K ctx Custom licence
    27.1KSept 2026gguf
    • Same category
    • Smaller (631M vs 771B)
    • Commercial use OK
    vs