metadata

tags:
  - vllm
  - vision
  - w4a16
license: apache-2.0
license_link: >-
  https://huggingface.co/datasets/choosealicense/licenses/blob/main/markdown/apache-2.0.md
language:
  - en
base_model: Qwen/Qwen2-VL-72B-Instruct
library_name: transformers

Qwen2-VL-72B-Instruct-quantized-w4a16

Model Overview

Model Architecture: Qwen/Qwen2-VL-72B-Instruct
- Input: Vision-Text
- Output: Text
Model Optimizations:
- Weight quantization: FP8
- Activation quantization: FP8
Release Date: 2/24/2025
Version: 1.0
Model Developers: Neural Magic

Quantized version of Qwen/Qwen2-VL-72B-Instruct.

Model Optimizations

This model was obtained by quantizing the weights of Qwen/Qwen2-VL-72B-Instruct to FP8 data type, ready for inference with vLLM >= 0.5.2.

Deployment

Use with vLLM

This model can be deployed efficiently using the vLLM backend, as shown in the example below.

from vllm.assets.image import ImageAsset
from vllm import LLM, SamplingParams

# prepare model
llm = LLM(
    model="neuralmagic/Qwen2-VL-72B-Instruct-quantized.w4a16",
    trust_remote_code=True,
    max_model_len=4096,
    max_num_seqs=2,
)

# prepare inputs
question = "What is the content of this image?"
inputs = {
    "prompt": f"<|user|>\n<|image_1|>\n{question}<|end|>\n<|assistant|>\n",
    "multi_modal_data": {
        "image": ImageAsset("cherry_blossom").pil_image.convert("RGB")
    },
}

# generate response
print("========== SAMPLE GENERATION ==============")
outputs = llm.generate(inputs, SamplingParams(temperature=0.2, max_tokens=64))
print(f"PROMPT  : {outputs[0].prompt}")
print(f"RESPONSE: {outputs[0].outputs[0].text}")
print("==========================================")

vLLM also supports OpenAI-compatible serving. See the documentation for more details.

Creation

This model was created with llm-compressor by running the code snippet below as part a multimodal announcement blog.

Model Creation Code

from transformers import AutoProcessor, Qwen2VLForConditionalGeneration

from llmcompressor.modifiers.quantization import QuantizationModifier
from llmcompressor.transformers import oneshot, wrap_hf_model_class

MODEL_ID = "Qwen/Qwen2-VL-72B-Instruct"

# Load model.
model_class = wrap_hf_model_class(Qwen2VLForConditionalGeneration)
model = model_class.from_pretrained(MODEL_ID, device_map="auto", torch_dtype="auto")
processor = AutoProcessor.from_pretrained(MODEL_ID)

# Configure the quantization algorithm and scheme.
# In this case, we:
#   * quantize the weights to fp8 with per channel via ptq
#   * quantize the activations to fp8 with dynamic per token
recipe = QuantizationModifier(
    targets="Linear",
    scheme="FP8_DYNAMIC",
    ignore=["re:.*lm_head", "re:visual.*"],
)

# Apply quantization and save to disk in compressed-tensors format.
SAVE_DIR = MODEL_ID.split("/")[1] + "-FP8-dynamic"
oneshot(model=model, recipe=recipe, output_dir=SAVE_DIR)
processor.save_pretrained(SAVE_DIR)

# Confirm generations of the quantized model look sane.
print("========== SAMPLE GENERATION ==============")
input_ids = processor(text="Hello my name is", return_tensors="pt").input_ids.to("cuda")
output = model.generate(input_ids, max_new_tokens=20)
print(processor.decode(output[0]))
print("==========================================")

Evaluation

Evaluation Commands

Accuracy

Inference Performance

This model achieves up to xxx speedup in single-stream deployment and up to xxx speedup in multi-stream asynchronous deployment, depending on hardware and use-case scenario. The following performance benchmarks were conducted with vLLM version 0.7.2, and GuideLLM.

Benchmarking Command

``` guidellm --model neuralmagic/Qwen2-VL-72B-Instruct-FP8-Dynamic --target "http://localhost:8000/v1" --data-type emulated --data prompt_tokens=,generated_tokens=,images=,width=,height= --max seconds 120 --backend aiohttp_server ```

Single-stream performance (measured with vLLM version 0.7.2)

			Document Visual Question Answering 1680W x 2240H 64/128		Visual Reasoning 640W x 480H 128/128		Image Captioning 480W x 360H 0/128
Hardware	Model	Average Cost Reduction	Latency (s)	QPD	Latency (s)th>	QPD	Latency (s)	QPD
A100x4	Qwen/Qwen2-VL-72B-Instruct		6.5	77	4.6	110	4.4	113
A100x2	neuralmagic/Qwen2-VL-72B-Instruct-quantized.w8a8	1.85	7.2	139	4.9	206	4.8	211
A100x1	neuralmagic/Qwen2-VL-72B-Instruct-quantized.w4a16	3.32	10.0	202	5.0	398	4.8	419
H100x4	Qwen/Qwen2-VL-72B-Instruct		4.4	66	3.0	97	2.9	99
H100x2	neuralmagic/Qwen2-VL-72B-Instruct-FP8-Dynamic	1.79	4.7	119	3.3	173	3.2	177
H100x1	neuralmagic/Qwen2-VL-72B-Instruct-quantized.w4a16	2.60	6.4	172	4.3	253	4.2	259

Multi-stream asynchronous performance (measured with vLLM version 0.7.2)

			Document Visual Question Answering 1680W x 2240H 64/128		Visual Reasoning 640W x 480H 128/128		Image Captioning 480W x 360H 0/128
Hardware	Model	Average Cost Reduction	Maximum throughput (QPS)	QPD	Maximum throughput (QPS)	QPD	Maximum throughput (QPS)	QPD
A100x4	Qwen/Qwen2-VL-72B-Instruct		0.3	169	1.1	538	1.2	595
	A100x2	neuralmagic/Qwen2-VL-72B-Instruct-quantized.w8a8	1.84	0.6	293	2.0	1021	2.3	1135
	A100x1	neuralmagic/Qwen2-VL-72B-Instruct-quantized.w4a16	2.73	0.6	314	3.2	1591	4.0	2019
H100x4	Qwen/Qwen2-VL-72B-Instruct		0.5	137	1.2	356	1.3	377
neuralmagic/Qwen2-VL-72B-Instruct-FP8-Dynamic	H100x2	1.70	0.8	236	2.2	623	2.4	669
neuralmagic/Qwen2-VL-72B-Instruct-quantized.w4a16	H100x1	2.35	1.3	350	3.3	910	3.6	994