Easy, advanced inference platform for large language models on Kubernetes

llmaz (pronounced /lima:z/), aims to provide a Production-Ready inference platform for large language models on Kubernetes. It closely integrates with the state-of-the-art inference backends to bring the leading-edge researches to cloud.

🌱 llmaz is alpha now, so API may change before graduating to Beta.

Concept

Feature Overview

• Easy of Use: People can quick deploy a LLM service with minimal configurations.
• Broad Backend Support: llmaz supports a wide range of advanced inference backends for different scenarios, like vLLM, SGLang, llama.cpp. Find the full list of supported backends here.
• Scaling Efficiency (WIP): llmaz works smoothly with autoscaling components like Cluster-Autoscaler or Karpenter to support elastic scenarios.
• Accelerator Fungibility (WIP): llmaz supports serving the same LLM with various accelerators to optimize cost and performance.
• SOTA Inference: llmaz supports the latest cutting-edge researches like Speculative Decoding or Splitwise(WIP) to run on Kubernetes.
• Various Model Providers: llmaz supports a wide range of model providers, such as HuggingFace, ModelScope, ObjectStores(aliyun OSS, more on the way). llmaz automatically handles the model loading requiring no effort from users.
• Multi-hosts Support: llmaz supports both single-host and multi-hosts scenarios with LWS from day 1.

Quick Start

Installation

Read the Installation for guidance.

Deploy

Here's a simplest sample for deploying facebook/opt-125m, all you need to do is to apply a Model and a Playground.

Please refer to examples to learn more.

Note: if your model needs Huggingface token for weight downloads, please run kubectl create secret generic modelhub-secret --from-literal=HF_TOKEN=<your token> ahead.

Model

apiVersion: llmaz.io/v1alpha1
kind: OpenModel
metadata:
  name: opt-125m
spec:
  familyName: opt
  source:
    modelHub:
      modelID: facebook/opt-125m
  inferenceFlavors:
  - name: t4 # GPU type
    requests:
      nvidia.com/gpu: 1

Inference Playground

apiVersion: inference.llmaz.io/v1alpha1
kind: Playground
metadata:
  name: opt-125m
spec:
  replicas: 1
  modelClaim:
    modelName: opt-125m

Test

Expose the service

kubectl port-forward pod/opt-125m-0 8080:8080

Get registered models

curl http://localhost:8080/v1/models

Request a query

curl http://localhost:8080/v1/completions \
-H "Content-Type: application/json" \
-d '{
    "model": "opt-125m",
    "prompt": "San Francisco is a",
    "max_tokens": 10,
    "temperature": 0
}'

Roadmap

• Gateway support for traffic routing
• Metrics support
• Serverless support for cloud-agnostic users
• CLI tool support
• Model training, fine tuning in the long-term

Project Structures

llmaz # root
├── llmaz # where the model loader logic locates
├── pkg # where the main logic for Kubernetes controllers locates

Contributions

🚀 All kinds of contributions are welcomed ! Please follow Contributing. Thanks to all these contributors.