Hugging Face's logo

Spaces:
hugging-apps
/
metaview
Running on Zero

App Files Community
metaview

Commit History

  1. Upload app.py with huggingface_hub Running

    multimodalart HF Staff
    Verified signature
    cfd4941

  2. Upload app.py with huggingface_hub

    multimodalart HF Staff
    Verified signature
    864a55e

  3. Upload app.py with huggingface_hub

    multimodalart HF Staff
    Verified signature
    5035fd0

  4. Simplify camera UI to coupled 2D pad + linked yaw/pitch sliders

    multimodalart HF Staff
    Verified signature
    fcf6034

  5. Backbone bf16 + fp32-recast hook at head boundary; keep DA3 heads fp32 to fit pack and stay dtype-consistent

    multimodalart HF Staff
    Verified signature
    e69128c

  6. Shrink only DA3 backbone to bf16 (keep heads fp32) to fit 76GB pack while preserving correctness

    multimodalart HF Staff
    Verified signature
    2c731e8

  7. Cast DA3 priors to bf16 but keep norm layers fp32 (fixes LayerNorm dtype error) to fit 76GB pack

    multimodalart HF Staff
    Verified signature
    772fc71

  8. Cast DA3 priors to bf16 so full resident set packs into the 76GB NVMe offload dir

    multimodalart HF Staff
    Verified signature
    23eacc5

  9. TEMP: probe /data-nvme exact bytes

    multimodalart HF Staff
    Verified signature
    ff68186

  10. TEMP: probe per-model resident byte sizes

    multimodalart HF Staff
    Verified signature
    7f88b2a

  11. Point ZeroGPU offload dir at /data-nvme block device

    multimodalart HF Staff
    Verified signature
    5ccae3b

  12. Redirect ZeroGPU offload dir to /dev/shm to fit the 78GB tensor pack; drop diagnostics

    multimodalart HF Staff
    Verified signature
    3ac437c

  13. TEMP: probe 90GB fallocate per filesystem

    multimodalart HF Staff
    Verified signature
    5d34583

  14. TEMP: probe O_DIRECT/fallocate support per filesystem

    multimodalart HF Staff
    Verified signature
    35a0b24

  15. TEMP: add df -h diagnostic for offload dir sizing

    multimodalart HF Staff
    Verified signature
    b58953f

  16. Adopt standard ZeroGPU recipe: .to('cuda') at module scope, drop _ensure_cuda

    multimodalart HF Staff
    Verified signature
    2c348e6

  17. Upload app.py with huggingface_hub

    multimodalart HF Staff
    Verified signature
    9c1909b

  18. Upload app.py with huggingface_hub

    multimodalart HF Staff
    Verified signature
    a29be96

  19. Upload app.py with huggingface_hub

    multimodalart HF Staff
    Verified signature
    36a1001

  20. Upload app.py with huggingface_hub

    multimodalart HF Staff
    Verified signature
    342c71d

  21. Upload app.py with huggingface_hub

    multimodalart HF Staff
    Verified signature
    7e44673

  22. Initial MetaView novel view synthesis demo

    multimodalart HF Staff
    Verified signature
    ccfee12

  23. initial commit

    multimodalart HF Staff
    Verified signature
    ce6e2fd