This is a BentoML example project, showing you how to serve and deploy open-source Large Language Models (LLMs) using LMDeploy, a toolkit for compressing, deploying, and serving LLMs.

See here for a full list of BentoML example projects.

💡 This example is served as a basis for advanced code customization, such as custom model, inference logic or LMDeploy options. For simple LLM hosting with OpenAI compatible endpoint without writing any code, see OpenLLM.

Prerequisites

• You have installed Python 3.8+ and pip. See the Python downloads page to learn more.
• You have a basic understanding of key concepts in BentoML, such as Services. We recommend you read Quickstart first.
• If you want to test the Service locally, you need a Nvidia GPU with at least 20G VRAM.
• This example uses Llama 3 8B Instruct. Make sure you have gained access to the model.
• (Optional) We recommend you create a virtual environment for dependency isolation for this project. See the Conda documentation or the Python documentation for details.

Install dependencies

bash git clone https://github.com/bentoml/BentoLMDeploy.git cd BentoLMDeploy/llama3.1-8b-instruct pip install -r requirements.txt

Download the model

Run the script to download Llama 3.

bash python import_model.py

Run the BentoML Service

We have defined a BentoML Service in service.py. Run bentoml serve in your project directory to start the Service.

```bash $ bentoml serve .

2024-05-04T17:24:01+0800 [INFO] [cli] Starting production HTTP BentoServer from "service:LMDeploy" listening on http://localhost:3000 (Press CTRL+C to quit) 2024-05-04 17:24:03,239 - lmdeploy - INFO - input backend=turbomind, backendconfig=TurbomindEngineConfig(modelname='meta-llama/Meta-Llama-3-8B-Instruct', modelformat='hf', tp=1, sessionlen=None, maxbatchsize=128, cach emaxentrycount=0.9, cacheblockseqlen=64, quantpolicy=0, ropescalingfactor=0.0, uselognattn=False, downloaddir=None, revision=None, maxprefilltokennum=8192, numtokensperiter=0, maxprefilliters=1) 2024-05-04 17:24:03,240 - lmdeploy - INFO - input chattemplateconfig=None 2024-05-04 17:24:03,339 - lmdeploy - INFO - updated chattemplateonfig=ChatTemplateConfig(modelname='llama3', system=None, metainstruction=None, eosys=None, user=None, eoh=None, assistant=None, eoa=None, separator=None, capability=None, stopwords=None) 2024-05-04 17:24:03,359 - lmdeploy - WARNING - modelsource: hfmodel 2024-05-04 17:24:03,359 - lmdeploy - WARNING - modelname is deprecated in TurbomindEngineConfig and has no effect Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained. Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained. 2024-05-04 17:24:03,727 - lmdeploy - WARNING - model_config:

... ```

The server is now active at http://localhost:3000. You can interact with it using the Swagger UI or in other different ways.

Deploy to BentoCloud

After the Service is ready, you can deploy the application to BentoCloud for better management and scalability. Sign up if you haven't got a BentoCloud account.

Make sure you have logged in to BentoCloud, then run the following command to deploy it.

bash bentoml deploy .

Once the application is up and running on BentoCloud, you can access it via the exposed URL.

Note: For custom deployment in your own infrastructure, use BentoML to generate an OCI-compliant image.