78d78cdf8a
* Add TorchAOHfQuantizer Summary: Enable loading torchao quantized model in huggingface. Test Plan: local test Reviewers: Subscribers: Tasks: Tags: * Fix a few issues * style * Added tests and addressed some comments about dtype conversion * fix torch_dtype warning message * fix tests * style * TorchAOConfig -> TorchAoConfig * enable offload + fix memory with multi-gpu * update torchao version requirement to 0.4.0 * better comments * add torch.compile to torchao README, add perf number link --------- Co-authored-by: Marc Sun <marc@huggingface.co>
1.8 KiB
Executable File
Quantization
Quantization techniques reduce memory and computational costs by representing weights and activations with lower-precision data types like 8-bit integers (int8). This enables loading larger models you normally wouldn't be able to fit into memory, and speeding up inference. Transformers supports the AWQ and GPTQ quantization algorithms and it supports 8-bit and 4-bit quantization with bitsandbytes.
Quantization techniques that aren't supported in Transformers can be added with the [HfQuantizer] class.
Learn how to quantize models in the Quantization guide.
QuantoConfig
autodoc QuantoConfig
AqlmConfig
autodoc AqlmConfig
AwqConfig
autodoc AwqConfig
EetqConfig
autodoc EetqConfig
GPTQConfig
autodoc GPTQConfig
BitsAndBytesConfig
autodoc BitsAndBytesConfig
HfQuantizer
autodoc quantizers.base.HfQuantizer
HqqConfig
autodoc HqqConfig
FbgemmFp8Config
autodoc FbgemmFp8Config
TorchAoConfig
autodoc TorchAoConfig