Hugging Face's logo

Spaces:
TiGa-RCE
/
embedding-quantization-cuda-control
Running on Zero

App Files Community
embedding-quantization-cuda-control / __pycache__ /app.cpython-313.pyc

Commit History

  1. Isolate GTE startup packing from completed 0.6B model

    TiGa-RCE
    Verified signature
    234931c

  2. Move GTE 1.5B CUDA placement to ZeroGPU startup

    TiGa-RCE
    Verified signature
    88534d9

  3. Add mounted 1.5B and 8B BF16 cross-runtime controls

    TiGa-RCE
    Verified signature
    7973967

  4. Lazy-load CUDA quantizer configuration inside GPU call

    TiGa-RCE
    Verified signature
    22338c1

  5. Add bounded bitsandbytes INT8 and NF4 controls

    TiGa-RCE
    Verified signature
    1a6347c

  6. Add bounded ZeroGPU BF16 CUDA control

    TiGa-RCE
    Verified signature
    bc3f111