Skip to content

Deployment

Instead of installing the CLI via pip, you may also use docker to run michaelf34/infinity. Make sure you mount your accelerator, i.e. install nvidia-docker and activate with --gpus all.

port=7997
model1=michaelfeil/bge-small-en-v1.5
model2=mixedbread-ai/mxbai-rerank-xsmall-v1
volume=$PWD/data

docker run -it --gpus all \
 -v $volume:/app/.cache \
 -p $port:$port \
 michaelf34/infinity:latest \
 v2 \
 --model-id $model1 \
 --model-id $model2 \
 --port $port
The cache path inside the docker container is set by the environment variable HF_HOME.

AMD Docker: Deploy on AMD Platform (MI200 Series and MI300 Series)

port=7997
model1=michaelfeil/bge-small-en-v1.5
model2=mixedbread-ai/mxbai-rerank-xsmall-v1
volume=$PWD/data

docker run -it \
  --cap-add=SYS_PTRACE \
  --security-opt seccomp=unconfined \
  --device=/dev/kfd \
  --device=/dev/dri \
  --group-add video \
  --network host \
  -v $volume:/app/.cache \
  -p $port:$port \
  michaelf34/infinity:latest-rocm \
  v2 \
  --model-id $model1 \
  --model-id $model2 \
  --port $port \
  --engine torch \
  --compile \
  --no-bettertransformer
The cache path at inside the docker container is set by the environment variable HF_HOME.

A deployment example for usage within are located at repo, including a Github Actions Pipeline.

The example is located at michaelfeil/infinity/tree/main/infra/modal.

The GPU and Modal-powered endpoint via this Github Pipeline is free to try out at infinity.modal.michaelfeil.eu, which is available at no cost.

Runpod.io - Serverless

There is a dedicated guide on how deploy via Runpod Serverless. Find out how to deploy it via this Repo: github.com/runpod-workers/worker-infinity-text-embeddings

Bento - BentoInfinity

Example repo for deployment via Bento: https://github.com/bentoml/BentoInfinity

dstack

dstack allows you to provision a VM instance on the cloud of your choice. Write a service configuration file as below for the deployment of BAAI/bge-small-en-v1.5 model wrapped in Infinity.

type: service

image: michaelf34/infinity:latest
env:
  - INFINITY_MODEL_ID=BAAI/bge-small-en-v1.5;BAAI/bge-reranker-base;
  - INFINITY_PORT=80
commands:
  - infinity_emb v2
port: 80

Then, simply run the following dstack command. After this, a prompt will appear to let you choose which VM instance to deploy the Infinity.

dstack run . -f infinity/serve.dstack.yml --gpu 16GB

For more detailed tutorial and general information about dstack, visit the official doc.

Docker with offline mode / models with custom pip packages

If you want to run infinity in a location without internet access, you can pre-download the model into the dockerfile. This is also the advised route to go, if you want to use infinity with models that require additional packages such as nomic-ai/nomic-embed-text-v1.5.

# clone the repo
git clone https://github.com/michaelfeil/infinity
git checkout tags/0.0.52
cd libs/infinity_emb
# build download stage using docker buildx buildkit.
docker buildx build --target=production-with-download \
--build-arg MODEL_NAME=michaelfeil/bge-small-en-v1.5 --build-arg ENGINE=torch \
-f Dockerfile -t infinity-model-small .
You can also set an argument EXTRA_PACKAGES if you require to install any extra packages. --build-arg EXTRA_PACKAGES="torch_geometric"

Rename and push it to your internal docker registry. 

docker tag infinity-model-small  myregistryhost:5000/myinfinity/infinity:0.0.52-small
docker push myregistryhost:5000/myinfinity/infinity:0.0.52-small

Note: You can also save a dockerfile direclty as .tar. This might come in handy if you do not have a shared internal docker registry in your nuclear facility, but still want to leverage the latest semantic search. https://docs.docker.com/reference/cli/docker/image/save/.

Remote image and audio URLs (Infinity v1)

INFINITY_MEDIA_URL_BLACKLIST is a semicolon-separated list of hostnames, host globs (e.g. *.example.com), and IP addresses/CIDRs to block. The default is:

localhost;*.localhost;0.0.0.0/8;10.0.0.0/8;100.64.0.0/10;127.0.0.0/8;
169.254.0.0/16;172.16.0.0/12;192.168.0.0/16;224.0.0.0/4;
::/128;::1/128;fc00::/7;fe80::/10;ff00::/8

Setting the variable replaces these defaults; an empty value disables blacklist matching. Full URLs/path prefixes are not supported: use their hostname instead. Only HTTP(S) URLs without credentials are accepted. Resolved connection addresses and every redirect are checked; environment proxies and .netrc are ignored. Public URLs/redirects and image/audio bytes or data URIs remain supported.

Remote downloads are streamed with a 128 MiB limit per item, including when Content-Length is absent. HTTP-compressed responses are rejected (requests ask for identity encoding). Decoded media and concurrent downloads can require more memory. Configure the limit in bytes with INFINITY_MEDIA_MAX_DOWNLOAD_BYTES (default: 134217728); the value must be a positive integer.