Qwen2-57B-A14B-Instruct-GPTQ-Int4

Introduction

Qwen2 is the new series of Qwen large language models. For Qwen2, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters, including a Mixture-of-Experts model. This repo contains the instruction-tuned 57B-A14B Mixture-of-Experts Qwen2 model.

Compared with the state-of-the-art opensource language models, including the previous released Qwen1.5, Qwen2 has generally surpassed most opensource models and demonstrated competitiveness against proprietary models across a series of benchmarks targeting for language understanding, language generation, multilingual capability, coding, mathematics, reasoning, etc.

Note: vLLM does not support the GPTQ version of Qwen2MoeForCausalLM currently.

Qwen2-57B-A14B-Instruct supports a context length of up to 65,536 tokens, enabling the processing of extensive inputs. However, since vLLM currently does not support this model (Qwen2-57B-A14B-Instruct-GPTQ-Int4), please refer to Qwen2-57B-A14B-Instruct.

For more details, please refer to our blog, GitHub, and Documentation.

Model Details

Qwen2 is a language model series including decoder language models of different model sizes. For each size, we release the base language model and the aligned chat model. It is based on the Transformer architecture with SwiGLU activation, attention QKV bias, group query attention, etc. Additionally, we have an improved tokenizer adaptive to multiple natural languages and codes.

Training details

We pretrained the models with a large amount of data, and we post-trained the models with both supervised finetuning and direct preference optimization.

Requirements

The code of Qwen2MoE has been in the latest Hugging face transformers and we advise you to install transformers>=4.40.0, or you might encounter the following error:

KeyError: 'qwen2_moe'

Quickstart

Here provides a code snippet with apply_chat_template to show you how to load the tokenizer and model and how to generate contents.

from transformers import AutoModelForCausalLM, AutoTokenizer
device = "cuda" # the device to load the model onto

model = AutoModelForCausalLM.from_pretrained(
    "Qwen/Qwen2-57B-A14B-Instruct-GPTQ-Int4",
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2-57B-A14B-Instruct-GPTQ-Int4")

prompt = "Give me a short introduction to large language model."
messages = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(device)

generated_ids = model.generate(
    model_inputs.input_ids,
    max_new_tokens=512
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]

Benchmark and Speed

To compare the generation performance between bfloat16 (bf16) and quantized models such as GPTQ-Int8, GPTQ-Int4, and AWQ, please consult our Benchmark of Quantized Models. This benchmark provides insights into how different quantization techniques affect model performance.

For those interested in understanding the inference speed and memory consumption when deploying these models with either transformer or vLLM, we have compiled an extensive Speed Benchmark.

Citation

If you find our work helpful, feel free to give us a cite.

@article{qwen2,
  title={Qwen2 Technical Report},
  year={2024}
}