[bugfix] fix gptq transformers>=5.0

[Jintao-Huang]

No description provided.

[gemini-code-assist]

Code Review

This pull request adds FP8 to the quantization support documentation and introduces a check for hf_device_map in the GPTQ quantization pipeline. Feedback was provided to avoid hardcoding the device to cuda:0 in the quantization logic and to fix inconsistent table alignment in the English documentation.

[gemini-code-assist]

Hardcoding the device to cuda:0 can lead to failures or incorrect behavior on multi-GPU systems or CPU-only environments. It is safer to use the model's current device when initializing the hf_device_map.

Suggested change

	if not hasattr(self.model, 'hf_device_map'):
	self.model.hf_device_map = {'': torch.device('cuda:0')}
	if not hasattr(self.model, 'hf_device_map'):
	self.model.hf_device_map = {'': self.model.device}