QuantLMs-4Bit-Unpacked
Collection
QuantLMs, unpacked to FP16 format - compatible with FP16 GEMMs. After unpacking, QuantLMs have the same architecture as LLaMa.
•
9 items
•
Updated
QuantLM, unpacked to FP16 format - compatible with FP16 GEMMs. After unpacking, QuantLM has the same architecture as LLaMa.
import transformers as tf, torch
model_name = "SpectraSuite/QuantLM_99M_4bit_Unpacked"
# Please adjust the temperature, repetition penalty, top_k, top_p and other sampling parameters according to your needs.
pipeline = tf.pipeline("text-generation", model=model_id, model_kwargs={"torch_dtype": torch.float16}, device_map="auto")
# These are base (pretrained) LLMs that are not instruction and chat tuned. You may need to adjust your prompt accordingly.
pipeline("Once upon a time")