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.
- User Friendly: People can quick deploy a LLM service with minimal configurations.
- High Performance: llmaz supports a wide range of advanced inference backends for high performance, 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 (WIP): llmaz supports the latest cutting-edge researches like Speculative Decoding or Splitwise to run on Kubernetes.
- Various Model Providers: llmaz automatically loads models from various providers, such as HuggingFace, ModelScope, ObjectStores(aliyun OSS, more on the way).
- Multi-hosts Support: llmaz supports both single-host and multi-hosts scenarios with LWS from day 1.
Read the Installation for guidance.
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.
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
apiVersion: inference.llmaz.io/v1alpha1
kind: Playground
metadata:
name: opt-125m
spec:
replicas: 1
modelClaim:
modelName: opt-125m
kubectl port-forward pod/opt-125m-0 8080:8080
curl http://localhost:8080/v1/models
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
}'
- Gateway support for traffic routing
- Serverless support for cloud-agnostic users
- CLI tool support
- Model training, fine tuning in the long-term
🚀 All kinds of contributions are welcomed ! Please follow Contributing. Thanks to all these contributors.