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 h2o-lightning-31b

120 open-weight models that do the same job as h2o-lightning-31b (31.3B 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
h2oai logo

h2o-lightning-31b

h2oai·Vision

≈21 GB

H2O-Lightning-31B is a 31B-parameter decision model from H2O.ai, built on google/gemma-4-31B-it. It answers typed decision questions about a record (a document, a ticket, a policy, a conversation), and about images that come with the record. It returns a probability for every option:

31.3Bh2o-model-non-commercial-licence-v1.1
690Oct 2026
--vram
any
--gpus
--quant
  1. 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 31.3B)
    • Commercial use OK
    vs
  2. CohereLabs logo

    aya-vision-32b

    CohereLabs·Vision

    ≈21 GB

    Cohere Labs Aya Vision 32B is an open weights research release of a 32-billion parameter model with advanced capabilities optimized for a variety of vision-language use cases, including OCR, captioning, visual reasoning, summarization, question answering, code, and more.

    33.1B8K ctx CC BY-NC 4.0
    299Mar 2025
    • Same category
    • Similar size (33.1B vs 31.3B)
    vs
  3. swiss-ai logo

    Apertus-v1.5-70B

    swiss-ai·Vision

    ≈46 GB

    Apertus 1.5 is a family of 8B and 70B parameter language models designed to advance the state of multilingual, multimodal, fully open, and transparent AI.

    72BApache 2.0
    7.2KJul 2026gguf
    • Same category
    • Larger (72B vs 31.3B)
    • Commercial use OK
    vs
  4. 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
    9.3KJan 2026gguf
    • Same category
    • Smaller (15.1B vs 31.3B)
    • Commercial use OK
    vs
  5. 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 31.3B)
    • Commercial use OK
    vs
  6. 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
    33.8KApr 2025gguf
    • Same category
    • Similar size (16.4B vs 31.3B)
    • Commercial use OK
    vs
  7. google logo

    paligemma2-28b-pt-896

    google·Vision

    ≈20 GB

    Transformers PaliGemma 2 28B weights, pre-trained with 896*896 input images and 512 token input/output text sequences. The model is available in the bfloat16 format for fine-tuning.

    27.7BGemma Terms of Use
    722Nov 2024
    • Same category
    • Similar size (27.7B vs 31.3B)
    vs
  8. 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 31.3B)
    • Commercial use OK
    vs
  9. moonshotai logo

    Kimi-VL-A3B-Thinking-2506

    moonshotai·Vision

    ≈13 GB

    This is an updated version of Kimi-VL-A3B-Thinking, with following improved abilities:

    16.4BA3B128K ctx MIT
    13.3KJun 2025gguf
    • Same category
    • Similar size (16.4B vs 31.3B)
    • Commercial use OK
    vs
  10. ByteDance logo

    UniVR-34B-Planning

    ByteDance·Vision

    ≈23 GB

    UniVR is the first framework that simultaneously learns complex reasoning, fine-grained physical dynamics, and long-term planning from pure visual demonstrations — without relying on dense image-text pairs or task-specific heuristics.

    34BCC BY 4.0
    3Jul 2026
    • Same category
    • Similar size (34B vs 31.3B)
    • Commercial use OK
    vs
  11. google logo

    gemma-3-12b-it

    google·Vision

    ≈11 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.

    12.2B128K ctx Gemma Terms of Use
    408.4KMar 2025ollama
    • Same category
    • Smaller (12.2B vs 31.3B)
    vs
  12. inclusionAI logo

    LLaDA-UI

    inclusionAI·Vision

    ≈11 GB

    LLaDA-UI is an MoE-based, block-wise diffusion vision-language GUI agent. It understands screenshots at their native aspect ratio and produces grounded coordinates or structured actions for mobile, desktop, and web interfaces.

    16.9BA2.1B8K ctx Unknown licence
    289Sept 2026
    • Same category
    • Similar size (16.9B vs 31.3B)
    vs
  13. llava-hf logo

    llava-v1.6-34b-hf

    llava-hf·Vision

    ≈22 GB

    The LLaVA-NeXT model was proposed in LLaVA-NeXT: Improved reasoning, OCR, and world knowledge by Haotian Liu, Chunyuan Li, Yuheng Li, Bo Li, Yuanhan Zhang, Sheng Shen, Yong Jae Lee.

    34.8B4K ctx Unknown licence
    2.5KMar 2024
    • Same category
    • Similar size (34.8B vs 31.3B)
    vs
  14. 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 31.3B)
    • Commercial use OK
    vs
  15. 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 31.3B)
    • Commercial use OK
    vs
  16. meta-llama logo

    Llama-Guard-4-12B

    meta-llama·Vision

    ≈9.1 GB

    Llama Guard 4 is a natively multimodal safety classifier with 12 billion parameters trained jointly on text and multiple images. Llama Guard 4 is a dense architecture pruned from the Llama 4 Scout pre-trained model and fine-tuned for content safety classification.

    12B10M ctx Llama 4 Community License
    186.7KApr 2025gguf
    • Same category
    • Smaller (12B vs 31.3B)
    vs
  17. 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 31.3B)
    • Commercial use OK
    vs
  18. Qwen logo

    Qwen3.5-122B-A10B

    Qwen·Vision

    ≈75 GB

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

    125BA10B256K ctx Apache 2.0
    737.5KFeb 2026ollama
    • Same category
    • Larger (125B vs 31.3B)
    • Commercial use OK
    vs
  19. moonshotai logo

    Kimi-VL-A3B-Thinking

    moonshotai·Vision

    ≈13 GB

    Please set a higher temperature for thinking model, especially when your problem requires relatively long thinking process. We have updated the following sample script for HF inference.

    16.4BA3B128K ctx MIT
    9.5KApr 2025
    • Same category
    • Similar size (16.4B vs 31.3B)
    • Commercial use OK
    vs
  20. NVIDIA Nemotron Nano v2 12B VL model enables multi-image reasoning and video understanding, along with strong document intelligence, visual Q&A and summarization capabilities.

    13.2BNVIDIA Open Model License
    15.3KOct 2025gguf
    • Same category
    • Smaller (13.2B vs 31.3B)
    • Commercial use OK
    vs
  21. OpenGVLab logo

    InternVL3_5-14B

    OpenGVLab·Vision

    ≈11 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.

    15.1B40K ctx Apache 2.0
    6.6KAug 2025gguf
    • Same category
    • Smaller (15.1B vs 31.3B)
    • Commercial use OK
    vs
  22. google logo

    translategemma-12b-it

    google·Vision

    ≈11 GB

    TranslateGemma is a family of lightweight, state-of-the-art open translation models from Google, based on the Gemma 3 family of models. TranslateGemma models are designed to handle translation tasks across 55 languages.

    13.2B128K ctx Gemma Terms of Use
    6.2KJan 2026ollama
    • Same category
    • Smaller (13.2B vs 31.3B)
    vs
  23. NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 is the full-precision release of NVIDIA Nemotron 3.5 Super VL, a vision-language model for reasoning, coding, tool use, document understanding, and video understanding.

    124BA12B1M ctx openmdw-1.1
    84.9KOct 2026gguf
    • Same category
    • Larger (124B vs 31.3B)
    • Commercial use OK
    vs
  24. 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 31.3B)
    • Commercial use OK
    • Widely used
    vs