README.md

metadata

license: apache-2.0
base_model:
  - rhymes-ai/Aria-sequential_mlp
  - rhymes-ai/Aria
pipeline_tag: image-text-to-text

Aria-sequential_mlp-FP8-dynamic

Warning: There is no inference code for transformers/vLLM yet!

FP8-Dynamic quantization from Aria-sequential_mlp made with LLM Compressor.

Generated with the following code:

from transformers import AutoProcessor, AutoModelForCausalLM
from llmcompressor.modifiers.quantization import QuantizationModifier
from llmcompressor.transformers import SparseAutoModelForCausalLM, oneshot

model_name = "rhymes-ai/Aria-sequential_mlp"

model = SparseAutoModelForCausalLM.from_pretrained(model_name, device_map="auto", torch_dtype="auto", trust_remote_code=True)
processor = AutoProcessor.from_pretrained(model_name, trust_remote_code=True)

recipe = QuantizationModifier(
    targets="Linear",
    scheme="FP8_DYNAMIC",
    ignore=["re:.*lm_head", "re:multi_modal_projector.*", "re:vision_tower.*"],
)

folder = model_name.split("/")[1] + "-FP8-Dynamic"
oneshot(model=model, recipe=recipe, output_dir=folder)
processor.save_pretrained(folder)