Deepspeed inference and infinity offload with bitsandbytes 4bit loaded model

[ocesp98]

Is it possible to use deepspeed inference with a 4/8-bit quantized model using bitsandbytes?

I use the bitsandbytes package like this:

nf4_config = BitsAndBytesConfig(
   load_in_4bit=True,
   bnb_4bit_quant_type="nf4",
   bnb_4bit_use_double_quant=True,
   bnb_4bit_compute_dtype=torch.bfloat16
)

model = AutoModelForCausalLM.from_pretrained('model_id', device_map="auto", quantization_config=nf4_config)

zero_config =  {
        "stage": 3,
        "offload_param": {
            "device": "cpu"
            }
        }

ds_model = deepspeed.init_inference(
            model=model,     
            mp_size=1,        
            zero=zero_config
        )

 pipe = pipeline("text-generation", model=model, tokenizer=tokenizer, max_new_tokens=self.max_new_tokens
            )

However, it throws an error:

ValueError: .to is not supported for 4-bit or 8-bit models. Please use the model as it is, since the model
has already been set to the correct devices and casted to the correct dtype.

The ultimate goal is to combine the quantization with deepspeed zero infinity offload in the hope to run a larger model that currently does not fit on my GPU.

[rubenCrayon]

Hi! Any update on this error? I replicated it using transformers==4.33.2 and deepspeed==0.10.3 and the response differ a bit but the root is the same as before:

ValueError: .half() is not supported for 4-bit or 8-bit models. Please use the model as it is, since the model has already been casted to the correct dtype.

[tjruwase]

@ocesp98, @rubenCrayon can you please try the updated zero-inference with 4bit quantization?
https://github.com/microsoft/DeepSpeedExamples/blob/master/inference/huggingface/zero_inference/README.md