Spaces:

TiGa-RCE
/

embedding-quantization-cuda-control

Running on Zero

App Files Files Community

embedding-quantization-cuda-control / app.py

Commit History

Add direct-load BF16 repeat comparison

647387b
verified

TiGa-RCE commited on 2 days ago

Add bounded larger-family CUDA quantizer controls

bcba9be
verified

TiGa-RCE commited on 2 days ago

Load 8B directly inside allocated ZeroGPU call

7740aa9
verified

TiGa-RCE commited on 2 days ago

Switch larger-family control to root-loaded Qwen3 8B

bd2dad5
verified

TiGa-RCE commited on 2 days ago

Disable unused KV cache for embedding controls

97e5e64
verified

TiGa-RCE commited on 2 days ago

Expose larger-family benchmark traceback

2fad4af
verified

TiGa-RCE commited on 2 days ago

Isolate GTE startup packing from completed 0.6B model

234931c
verified

TiGa-RCE commited on 2 days ago

Move GTE 1.5B CUDA placement to ZeroGPU startup

88534d9
verified

TiGa-RCE commited on 2 days ago

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

7973967
verified

TiGa-RCE commited on 2 days ago

Lazy-load CUDA quantizer configuration inside GPU call

22338c1
verified

TiGa-RCE commited on 2 days ago

Add bounded bitsandbytes INT8 and NF4 controls

1a6347c
verified

TiGa-RCE commited on 2 days ago

Add bounded ZeroGPU BF16 CUDA control

bc3f111
verified

TiGa-RCE commited on 2 days ago