Docker

Weaviate supports deployment with Docker.

You can run Weaviate with default settings from a command line, or customize your configuration by creating your own docker-compose.yml file.

To run Weaviate with Docker using default settings, run this command from from your shell:

docker run -p 8080:8080 -p 50051:50051 cr.weaviate.io/semitechnologies/weaviate:1.38.2

The command sets the following default environment variables in the container:

• PERSISTENCE_DATA_PATH defaults to ./data
• AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED defaults to true.
• QUERY_DEFAULTS_LIMIT defaults to 10.

You can customize your Weaviate configuration by creating a docker-compose.yml file. Start from our sample Docker Compose file, or use the interactive Configurator to generate a docker-compose.yml file.

This starter Docker Compose file allows:

• Use of any API-based model provider integrations (e.g. OpenAI, Cohere, Google, and Anthropic).
  • This includes the relevant embedding model, generative, and reranker integrations.
• Searching pre-vectorized data (without a vectorizer).
• Mounts a persistent volume called weaviate_data to /var/lib/weaviate in the container to store data.

---
services:
  weaviate:
    command:
    - --host
    - 0.0.0.0
    - --port
    - '8080'
    - --scheme
    - http
    image: cr.weaviate.io/semitechnologies/weaviate:1.38.2
    ports:
    - 8080:8080
    - 50051:50051
    volumes:
    - weaviate_data:/var/lib/weaviate
    restart: on-failure:0
    environment:
      QUERY_DEFAULTS_LIMIT: 25
      AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED: 'true'
      PERSISTENCE_DATA_PATH: '/var/lib/weaviate'
      CLUSTER_HOSTNAME: 'node1'
volumes:
  weaviate_data:
...

---
services:
  weaviate:
    command:
    - --host
    - 0.0.0.0
    - --port
    - '8080'
    - --scheme
    - http
    image: cr.weaviate.io/semitechnologies/weaviate:1.38.2
    ports:
    - 8080:8080
    - 50051:50051
    volumes:
    - weaviate_data:/var/lib/weaviate
    restart: on-failure:0
    environment:
      QUERY_DEFAULTS_LIMIT: 25
      PERSISTENCE_DATA_PATH: '/var/lib/weaviate'
      CLUSTER_HOSTNAME: 'node1'
      AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED: 'false'
      AUTHENTICATION_APIKEY_ENABLED: 'true'
      AUTHENTICATION_APIKEY_ALLOWED_KEYS: 'user-a-key,user-b-key'
      AUTHENTICATION_APIKEY_USERS: 'user-a,user-b'
      AUTHORIZATION_ENABLE_RBAC: 'true'
      AUTHORIZATION_RBAC_ROOT_USERS: 'user-a'
volumes:
  weaviate_data:
...

Edit the docker-compose.yml file to suit your needs. You can add or remove environment variables, change the port mappings, or add additional model provider integrations, such as Ollama, or Hugging Face Transformers.

To start your Weaviate instance, run this command from your shell:

docker compose up -d

To make Weaviate accessible over both HTTP and gRPC, you need to expose their respective ports (by default 8080 for HTTP and 50051 for gRPC). To access these services via a domain, configure your reverse proxy to forward traffic to both ports. The recommended approach is to use a subdomain prefixed with grpc- for gRPC traffic, which ensures compatibility with most Weaviate clients.

For example, if your domain is weaviate.example.com, configure your reverse proxy as follows:

• weaviate.example.com → localhost:8080 (HTTP)
• grpc-weaviate.example.com → localhost:50051 (gRPC, typically using h2c)

The Configurator (experimental) can generate a docker-compose.yml file for you. Use the Configurator to select specific Weaviate modules, including vectorizers that run locally (i.e. text2vec-transformers, or multi2vec-clip)

You can use environment variables to control your Weaviate setup, authentication and authorization, module settings, and data storage settings.

Here are some examples of how to configure docker-compose.yml.

We recommended setting a persistent volume to avoid data loss as well as to improve reading and writing speeds.

Make sure to run docker compose down when shutting down. This writes all the files from memory to disk.

With named volume

services:
  weaviate:
    volumes:
        - weaviate_data:/var/lib/weaviate
    # etc

volumes:
    weaviate_data:

After running a docker compose up -d, Docker will create a named volume weaviate_data and mount it to the PERSISTENCE_DATA_PATH inside the container.

With host binding

services:
  weaviate:
    volumes:
      - /var/weaviate:/var/lib/weaviate
    # etc

After running a docker compose up -d, Docker will mount /var/weaviate on the host to the PERSISTENCE_DATA_PATH inside the container.

An example Docker Compose setup for Weaviate without any modules can be found below. In this case, no model inference is performed at either import or search time. You will need to provide your own vectors (e.g. from an outside ML model) at import and search time:

services:
  weaviate:
    image: cr.weaviate.io/semitechnologies/weaviate:1.38.2
    ports:
    - 8080:8080
    - 50051:50051
    restart: on-failure:0
    environment:
      QUERY_DEFAULTS_LIMIT: 25
      AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED: 'true'
      PERSISTENCE_DATA_PATH: '/var/lib/weaviate'
      CLUSTER_HOSTNAME: 'node1'

An example Docker Compose file with the transformers model sentence-transformers/multi-qa-MiniLM-L6-cos-v1 is:

services:
  weaviate:
    image: cr.weaviate.io/semitechnologies/weaviate:1.38.2
    restart: on-failure:0
    ports:
    - 8080:8080
    - 50051:50051
    environment:
      QUERY_DEFAULTS_LIMIT: 20
      AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED: 'true'
      PERSISTENCE_DATA_PATH: "./data"
      DEFAULT_VECTORIZER_MODULE: text2vec-transformers
      ENABLE_MODULES: text2vec-transformers
      TRANSFORMERS_INFERENCE_API: http://text2vec-transformers:8080
      CLUSTER_HOSTNAME: 'node1'
  text2vec-transformers:
    image: cr.weaviate.io/semitechnologies/transformers-inference:sentence-transformers-multi-qa-MiniLM-L6-cos-v1
    environment:
      ENABLE_CUDA: 0 # set to 1 to enable
      # NVIDIA_VISIBLE_DEVICES: all # enable if running with CUDA

Note that transformer models are neural networks built to run on GPUs. Running Weaviate with the text2vec-transformers module and without GPU is possible, but it will be slower. Enable CUDA with ENABLE_CUDA=1 if you have a GPU available.

For more information on how to set up the environment with the text2vec-transformers integration, see this page.

The text2vec-transformers module requires at least Weaviate version v1.2.0.

To run an unreleased version of Weaviate, edit your configuration file to use the unreleased image instead of a generally available image. The GitHub releases page lists generally available and release candidate builds.

For example, to run a Docker image for a release candidate, edit your docker-config.yaml to import the release candidate image.

image: cr.weaviate.io/semitechnologies/weaviate:1.34.0-rc.1

To configure Weaviate to use multiple host nodes, follow these steps:

• Configure one node as a "founding" member
• Set the CLUSTER_JOIN variable for the other nodes in the cluster.
• Set the CLUSTER_GOSSIP_BIND_PORT for each node.
• Set the CLUSTER_DATA_BIND_PORT for each node.
• Set the RAFT_JOIN each node.
• Set the RAFT_BOOTSTRAP_EXPECT for each node with the number of voters.
• Optionally, set the hostname for each node using CLUSTER_HOSTNAME.

(Read more about horizontal replication in Weaviate.)

So, the Docker Compose file includes environment variables for the "founding" member that look like this:

  weaviate-node-1:  # Founding member service name
    ...  # truncated for brevity
    environment:
      CLUSTER_HOSTNAME: 'node1'
      CLUSTER_GOSSIP_BIND_PORT: '7100'
      CLUSTER_DATA_BIND_PORT: '7101'
      RAFT_JOIN: 'node1,node2,node3'
      RAFT_BOOTSTRAP_EXPECT: 3

And the other members' configurations may look like this:

  weaviate-node-2:
    ...  # truncated for brevity
    environment:
      CLUSTER_HOSTNAME: 'node2'
      CLUSTER_GOSSIP_BIND_PORT: '7102'
      CLUSTER_DATA_BIND_PORT: '7103'
      CLUSTER_JOIN: 'weaviate-node-1:7100'  # This must be the service name of the "founding" member node.
      RAFT_JOIN: 'node1,node2,node3'
      RAFT_BOOTSTRAP_EXPECT: 3

Below is an example configuration for a 3-node setup. You may be able to test replication examples locally using this configuration.

services:
  weaviate-node-1:
    init: true
    command:
    - --host
    - 0.0.0.0
    - --port
    - '8080'
    - --scheme
    - http
    image: cr.weaviate.io/semitechnologies/weaviate:1.38.2
    ports:
    - 8080:8080
    - 6060:6060
    - 50051:50051
    restart: on-failure:0
    volumes:
      - ./data-node-1:/var/lib/weaviate
    environment:
      LOG_LEVEL: 'debug'
      QUERY_DEFAULTS_LIMIT: 25
      AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED: 'true'
      PERSISTENCE_DATA_PATH: '/var/lib/weaviate'
      CLUSTER_HOSTNAME: 'node1'
      CLUSTER_GOSSIP_BIND_PORT: '7100'
      CLUSTER_DATA_BIND_PORT: '7101'
      RAFT_JOIN: 'node1,node2,node3'
      RAFT_BOOTSTRAP_EXPECT: 3

  weaviate-node-2:
    init: true
    command:
    - --host
    - 0.0.0.0
    - --port
    - '8080'
    - --scheme
    - http
    image: cr.weaviate.io/semitechnologies/weaviate:1.38.2
    ports:
    - 8081:8080
    - 6061:6060
    - 50052:50051
    restart: on-failure:0
    volumes:
      - ./data-node-2:/var/lib/weaviate
    environment:
      LOG_LEVEL: 'debug'
      QUERY_DEFAULTS_LIMIT: 25
      AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED: 'true'
      PERSISTENCE_DATA_PATH: '/var/lib/weaviate'
      CLUSTER_HOSTNAME: 'node2'
      CLUSTER_GOSSIP_BIND_PORT: '7102'
      CLUSTER_DATA_BIND_PORT: '7103'
      CLUSTER_JOIN: 'weaviate-node-1:7100'
      RAFT_JOIN: 'node1,node2,node3'
      RAFT_BOOTSTRAP_EXPECT: 3

  weaviate-node-3:
    init: true
    command:
    - --host
    - 0.0.0.0
    - --port
    - '8080'
    - --scheme
    - http
    image: cr.weaviate.io/semitechnologies/weaviate:1.38.2
    ports:
    - 8082:8080
    - 6062:6060
    - 50053:50051
    restart: on-failure:0
    volumes:
      - ./data-node-3:/var/lib/weaviate
    environment:
      LOG_LEVEL: 'debug'
      QUERY_DEFAULTS_LIMIT: 25
      AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED: 'true'
      PERSISTENCE_DATA_PATH: '/var/lib/weaviate'
      CLUSTER_HOSTNAME: 'node3'
      CLUSTER_GOSSIP_BIND_PORT: '7104'
      CLUSTER_DATA_BIND_PORT: '7105'
      CLUSTER_JOIN: 'weaviate-node-1:7100'
      RAFT_JOIN: 'node1,node2,node3'
      RAFT_BOOTSTRAP_EXPECT: 3

The output of docker compose up is quite verbose as it attaches to the logs of all containers.

You can attach the logs only to Weaviate itself, for example, by running the following command instead of docker compose up:

# Run Docker Compose
docker compose up -d && docker compose logs -f weaviate

Alternatively you can run docker compose entirely detached with docker compose up -d and then poll {bindaddress}:{port}/v1/meta until you receive a status 200 OK.

In some systems, the cluster hostname may change over time. This is known to create issues with a single-node Weaviate deployment. To avoid this, set the CLUSTER_HOSTNAME environment variable in your docker-compose.yml file to the cluster hostname.

---
services:
  weaviate:
    # ...
    environment:
      CLUSTER_HOSTNAME: 'node1'
...

• If you are new to Docker, see Docker Introduction for Weaviate Users.

Have a question or feedback? Here's how to reach us.