Marc Sun
f62dc9bf2c
* gguf non quantized
* add support for q_k
* kernels and config
* better integration with kernels
* small update
* fix
* fix
* support for ggml-attn
* update
* fix
* update
* fix
* kernels gated delta
* Revert "kernels gated delta"
This reverts commit b963ea10e5.
* kernels gated delta agin
* input permutation
* style
* update integration
* up
* docs and bunch of fixes
* tokenizer for qwen3.5
* fix ci and doc
* cleaner integration
* up
* fix
* update
* fix
* up
* serve fix
* doc
* leftover stuff
* minor fix
* norm ! 11% speed increase
* up
* correct path !
* update
* less verbose
2.4 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
VptqConfig
autodoc VptqConfig
AwqConfig
autodoc AwqConfig
EetqConfig
autodoc EetqConfig
GPTQConfig
autodoc GPTQConfig
BitsAndBytesConfig
autodoc BitsAndBytesConfig
HfQuantizer
autodoc quantizers.base.HfQuantizer
HiggsConfig
autodoc HiggsConfig
HqqConfig
autodoc HqqConfig
MetalConfig
autodoc MetalConfig
Mxfp4Config
autodoc Mxfp4Config
NVFP4Config
autodoc NVFP4Config
FbgemmFp8Config
autodoc FbgemmFp8Config
CompressedTensorsConfig
autodoc CompressedTensorsConfig
TorchAoConfig
autodoc TorchAoConfig
BitNetQuantConfig
autodoc BitNetQuantConfig
SpQRConfig
autodoc SpQRConfig
FineGrainedFP8Config
autodoc FineGrainedFP8Config
QuarkConfig
autodoc QuarkConfig
FourOverSixConfig
autodoc FourOverSixConfig
FPQuantConfig
autodoc FPQuantConfig
AutoRoundConfig
autodoc AutoRoundConfig
SinqConfig
autodoc SinqConfig
GgufConfig
autodoc GgufConfig