​
Run with pip install

1

Run `pip install`

To install Letta, run:

pip install letta
2

Configure Letta

Run letta configure or letta quickstart to configure Letta.

To customize the configuration (e.g. the LLM or embedding model to use), edit the ~/.letta/config file.

~/.letta/config
[model]
model = gpt-4
model_endpoint = https://api.openai.com/v1
model_endpoint_type = openai
context_window = 8192

[embedding]
embedding_endpoint_type = openai
embedding_endpoint = https://api.openai.com/v1
embedding_model = text-embedding-ada-002
embedding_dim = 1536
embedding_chunk_size = 300
3

Run the Letta server

To run the Letta server, run:

letta run [--debug]

You can now access the ADE (in your browser) and REST API server at http://localhost:8283.

​
Run with Docker

1

Download the docker container

To run the docker container, first clone the reposity for pull the contianer.

git clone https://github.com/cpacker/MemGPT
2

Set enviornment variables

Either set the enviornment variables in your shell on in a .env file.

.env

# To use OpenAI 
OPENAI_API_KEY=...  

# To use with Ollama 
LETTA_LLM_ENDPOINT=http://host.docker.internal:11434
LETTA_LLM_ENDPOINT_TYPE=ollama
LETTA_LLM_MODEL=dolphin2.2-mistral:7b-q6_K
LETTA_LLM_CONTEXT_WINDOW=8192
LETTA_EMBEDDING_ENDPOINT=http://host.docker.internal:11434
LETTA_EMBEDDING_ENDPOINT_TYPE=ollama
LETTA_EMBEDDING_MODEL=mxbai-embed-large
LETTA_EMBEDDING_DIM=512
3

View and modify the `compose.yaml` file

You can view and modify the compose.yaml file.

docker compose up

You can now access the ADE (in your browser) and REST API server at http://localhost:8083.

QuickstartOpenAI