--- license: apache-2.0 pipeline_tag: text-generation tags: - text-generation-inference ---
English | [简体中文](README_zh-CN.md)

👋 join us on Twitter, Discord and WeChat

# W4A16 LLM Model Deployment LMDeploy supports LLM model inference of 4-bit weight, with the minimum requirement for NVIDIA graphics cards being sm80. Before proceeding with the inference, please ensure that lmdeploy(>=v0.0.14) is installed. ```shell pip install 'lmdeploy>=0.0.14' ``` ## 4-bit LLM model Inference You can download the pre-quantized 4-bit weight models from LMDeploy's [model zoo](https://huggingface.co/lmdeploy) and conduct inference using the following command. Alternatively, you can quantize 16-bit weights to 4-bit weights following the ["4-bit Weight Quantization"](#4-bit-weight-quantization) section, and then perform inference as per the below instructions. Take the 4-bit Qwen-Chat-14B model from the model zoo as an example: ```shell git-lfs install git clone https://huggingface.co/lmdeploy/qwen-chat-14b-4bit ``` As demonstrated in the command below, first convert the model's layout using `turbomind.deploy`, and then you can interact with the AI assistant in the terminal ```shell ## Convert the model's layout and store it in the default path, ./workspace. lmdeploy convert \ --model-name qwen-14b \ --model-path ./qwen-chat-14b-4bit \ --model-format awq \ --group-size 128 ## inference lmdeploy chat ./workspace ``` ## Serve with gradio If you wish to interact with the model via web ui, please initiate the gradio server as indicated below: ```shell lmdeploy serve gradio ./workspace --server_name {ip_addr} --server_port {port} ``` Subsequently, you can open the website `http://{ip_addr}:{port}` in your browser and interact with the model ## Inference Performance We benchmarked the Llama 2 7B and 13B with 4-bit quantization on NVIDIA GeForce RTX 4090 using [profile_generation.py](https://github.com/InternLM/lmdeploy/blob/main/benchmark/profile_generation.py). And we measure the token generation throughput (tokens/s) by setting a single prompt token and generating 512 tokens. All the results are measured for single batch inference. | model | llm-awq | mlc-llm | turbomind | | ----------- | ------- | ------- | --------- | | Llama 2 7B | 112.9 | 159.4 | 206.4 | | Llama 2 13B | N/A | 90.7 | 115.8 | ```shell pip install nvidia-ml-py ``` ```bash python profile_generation.py \ --model-path /path/to/your/model \ --concurrency 1 8 --prompt-tokens 0 512 --completion-tokens 2048 512 ``` ## 4-bit Weight Quantization It includes two steps: - generate quantization parameter - quantize model according to the parameter ### Step 1: Generate Quantization Parameter ```shell lmdeploy lite calibrate \ --model $HF_MODEL \ --calib_dataset 'c4' \ # Calibration dataset, supports c4, ptb, wikitext2, pileval --calib_samples 128 \ # Number of samples in the calibration set, if memory is insufficient, you can appropriately reduce this --calib_seqlen 2048 \ # Length of a single piece of text, if memory is insufficient, you can appropriately reduce this --work_dir $WORK_DIR \ # Folder storing Pytorch format quantization statistics parameters and post-quantization weight ``` ### Step2: Quantize Weights LMDeploy employs AWQ algorithm for model weight quantization. ```shell lmdeploy lite auto_awq \ --model $HF_MODEL \ --w_bits 4 \ # Bit number for weight quantization --w_sym False \ # Whether to use symmetric quantization for weights --w_group_size 128 \ # Group size for weight quantization statistics --work_dir $WORK_DIR \ # Directory saving quantization parameters from Step 1 ``` After the quantization is complete, the quantized model is saved to `$WORK_DIR`. Then you can proceed with model inference according to the instructions in the ["4-Bit Weight Model Inference"](#4-bit-llm-model-inference) section.