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 LongCat-2.5

120 open-weight models that do the same job as LongCat-2.5 (1.78T 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
meituan-longcat logo

LongCat-2.5

meituan-longcat·Vision

≈1070 GB

LongCat-2.5 is a native multimodal MoE model for long-horizon agentic tasks, with 1.6 trillion total parameters and approximately 48 billion activated per token.

1.78T512K ctx MIT
2.5KNov 2026
--vram
any
--gpus
--quant
  1. Qwen logo

    Qwen3.5-0.8B-Base

    Qwen·Vision

    ≈1.4 GB

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

    873M256K ctx Apache 2.0
    867.8KFeb 2026gguf
    • Same category
    • Smaller (873M vs 1.78T)
    • Commercial use OK
    vs
  2. cantina-security logo

    apex-flash-1-abliterated

    cantina-security·Vision

    ≈213 GB

    apex-flash-1-abliterated is an experimental derivative of apex-flash-1, Cantina Security's open-weights security model developed in partnership with Yeta (@yetalabs on X).

    321BA18.3B1M ctx MIT
    710Sept 2026gguf
    • Same category
    • Smaller (321B vs 1.78T)
    • Commercial use OK
    vs
  3. 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
    767.6KFeb 2026ollama
    • Same category
    • Smaller (125B vs 1.78T)
    • Commercial use OK
    vs
  4. huihui-ai logo≈19 GB

    This is an uncensored version of Qwen/Qwen3.8-27B created with abliteration (see remove-refusals-with-transformers to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.

    27.8B256K ctx Apache 2.0
    140.7KAug 2026gguf
    • Same category
    • Smaller (27.8B vs 1.78T)
    • Commercial use OK
    vs
  5. 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
    155.7KAug 2026ollama
    • Same category
    • Smaller (29.8B vs 1.78T)
    • Commercial use OK
    vs
  6. 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
    149.4KAug 2026
    • Same category
    • Smaller (2.6B vs 1.78T)
    • Commercial use OK
    vs
  7. 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
    503.2KFeb 2026gguf
    • Same category
    • Smaller (4.7B vs 1.78T)
    • Commercial use OK
    vs
  8. 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
    338.6KApr 2026gguf
    • Same category
    • Smaller (9.7B vs 1.78T)
    • Commercial use OK
    vs
  9. 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
    973Jun 2025
    • Same category
    • Smaller (424B vs 1.78T)
    • Commercial use OK
    vs
  10. 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
    78KSept 2026ollama
    • Same category
    • Smaller (27.4B vs 1.78T)
    • Commercial use OK
    vs
  11. 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
    146Oct 2026
    • Same category
    vs
  12. 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 1.78T)
    • Commercial use OK
    vs
  13. 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
    74.6KOct 2026gguf
    • Same category
    • Smaller (124B vs 1.78T)
    • Commercial use OK
    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
    64.7KOct 2026gguf
    • Same category
    • Smaller (9.4B vs 1.78T)
    • Commercial use OK
    vs
  15. Qwen logo≈142 GB

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

    236BA22B256K ctx Apache 2.0
    81KSept 2025gguf
    • Same category
    • Smaller (236B vs 1.78T)
    • Commercial use OK
    vs
  16. 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
    327.3KFeb 2026gguf
    • Same category
    • Smaller (2.3B vs 1.78T)
    • Commercial use OK
    vs
  17. 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
    496.1KDec 2025gguf
    • Same category
    • Smaller (2.4B vs 1.78T)
    • Commercial use OK
    vs
  18. 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
    88.5KJul 2026gguf
    • Same category
    • Smaller (10B vs 1.78T)
    • Commercial use OK
    vs
  19. 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
    188KApr 2026ollama
    • Same category
    • Smaller (1.3B vs 1.78T)
    • Commercial use OK
    vs
  20. 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
    265KFeb 2026gguf
    • Same category
    • Smaller (9.7B vs 1.78T)
    • Commercial use OK
    vs
  21. 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
    5.4MJan 2025ollama
    • Same category
    • Smaller (8.3B vs 1.78T)
    • Commercial use OK
    • Widely used
    vs
  22. 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
    34.9KOct 2026gguf
    • Same category
    • Smaller (4.8B vs 1.78T)
    • Commercial use OK
    vs
  23. 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
    457Jun 2025
    • Same category
    • Smaller (424B vs 1.78T)
    • Commercial use OK
    vs
  24. 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
    33.9KSept 2026ollama
    • Same category
    • Smaller (9.4B vs 1.78T)
    • Commercial use OK
    vs