View example on GitHub

This guide walks through deploying Mistral-7B, a powerful large language model (LLM), using Truss. You’ll configure the model, set up inference, allocate resources, and deploy it as an API endpoint.

Set up your model

Start by importing the necessary libraries:

model/model.py

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

Specify the Hugging Face model checkpoint:

model/model.py

CHECKPOINT = "mistralai/Mistral-7B-v0.1"

Define the model class

Create a Model class that loads Mistral-7B and its tokenizer when the server starts:

model/model.py

class Model:
    def __init__(self, **kwargs) -> None:
        self.tokenizer = None
        self.model = None

    def load(self):
        self.model = AutoModelForCausalLM.from_pretrained(
            CHECKPOINT, torch_dtype=torch.float16, device_map="auto"
        )
        self.tokenizer = AutoTokenizer.from_pretrained(CHECKPOINT)

Implement inference

The predict function handles inference by tokenizing input, generating text, and decoding the output.

model/model.py

    def predict(self, request: dict):
        prompt = request.pop("prompt")
        generate_args = {
            "max_new_tokens": request.get("max_new_tokens", 128),
            "temperature": request.get("temperature", 1.0),
            "top_p": request.get("top_p", 0.95),
            "top_k": request.get("top_k", 50),
            "repetition_penalty": 1.0,
            "use_cache": True,
            "do_sample": True,
            "eos_token_id": self.tokenizer.eos_token_id,
            "pad_token_id": self.tokenizer.pad_token_id,
        }

        input_ids = self.tokenizer(prompt, return_tensors="pt").input_ids.cuda()

        with torch.no_grad():
            output = self.model.generate(input_ids=input_ids, **generate_args)
            return self.tokenizer.decode(output[0])

Configure your deployment

Define dependencies

Specify the necessary Python packages in config.yaml:

config.yaml

model_name: Mistral 7B
python_version: py311
requirements:
  - transformers==4.42.3
  - sentencepiece==0.1.99
  - accelerate==0.23.0
  - torch==2.0.1
  - numpy==1.26.4

Allocate compute resources

Mistral-7B requires an NVIDIA A10G GPU for efficient inference:

config.yaml

resources:
  accelerator: A10G
  use_gpu: true

Deploy the model

Push your model to Baseten:

$ baseten model push

Once deployed, call the model using the Baseten CLI:

$ baseten model predict --model-id <model-id> --data '{"prompt": "What is a large language model?"}'

Or send a request to the API endpoint:

import requests

response = requests.post(
    "https://model-{yourmodelid}.api.baseten.co/production/predict",
    headers={"Authorization": "Bearer EMPTY"},
    json={"prompt": "Explain quantum computing in simple terms"}
)

print(response.json())

You can also use the Truss CLI: truss push to deploy and truss predict --published -d '{"prompt": "..."}' to call the model.

Check for optimized engine support

For optimized performance we have open-source and Baseten optimized engines, such as Baseten’s TensorRT-LLM, Baseten-Embeddings-Inference, vLLM and SGLang.

Examples

Models

Engines

Custom Docker servers

Custom Python models

Chains

Build and deploy an LLM

View example on GitHub

Set up your model

Define the model class

Implement inference

Configure your deployment

Define dependencies

Allocate compute resources

Deploy the model

Check for optimized engine support

View example on GitHub

​Set up your model

​Define the model class

​Implement inference

​Configure your deployment

​Define dependencies

​Allocate compute resources

​Deploy the model

​Check for optimized engine support

Set up your model

Define the model class

Implement inference

Configure your deployment

Define dependencies

Allocate compute resources

Deploy the model

Check for optimized engine support