1curl -fsSL https://ollama.com/install.sh | sh

 1➜ ollama
 2Usage:
 3  ollama [flags]
 4  ollama [command]
 5
 6Available Commands:
 7  serve       Start ollama
 8  create      Create a model from a Modelfile
 9  show        Show information for a model
10  run         Run a model
11  pull        Pull a model from a registry
12  push        Push a model to a registry
13  list        List models
14  cp          Copy a model
15  rm          Remove a model
16  help        Help about any command
17
18Flags:
19  -h, --help      help for ollama
20  -v, --version   Show version information
21
22Use "ollama [command] --help" for more information about a command.

1conda create --name ollama # create a new env
2conda activate ollama
3
4# install packages 
5
6conda install conda-forge::chromadb
7conda install pysqlite3 
8conda install pip
9pip3 install ollama # have to use pip to install ollama py library 

 1import ollama
 2import chromadb
 3
 4documents = [
 5  "Llamas are members of the camelid family meaning they're pretty closely related to vicuñas and camels",
 6  "Llamas were first domesticated and used as pack animals 4,000 to 5,000 years ago in the Peruvian highlands",
 7  "Llamas can grow as much as 6 feet tall though the average llama between 5 feet 6 inches and 5 feet 9 inches tall",
 8  "Llamas weigh between 280 and 450 pounds and can carry 25 to 30 percent of their body weight",
 9  "Llamas are vegetarians and have very efficient digestive systems",
10  "Llamas live to be about 20 years old, though some only live for 15 years and others live to be 30 years old",
11]
12
13client = chromadb.Client()
14collection = client.create_collection(name="docs")
15
16# store each document in a vector embedding database
17for i, d in enumerate(documents):
18  response = ollama.embeddings(model="mxbai-embed-large", prompt=d)
19  embedding = response["embedding"]
20  collection.add(
21    ids=[str(i)],
22    embeddings=[embedding],
23    documents=[d]
24  )

 1# an example prompt
 2prompt = "What animals are llamas related to?"
 3
 4# generate an embedding for the prompt and retrieve the most relevant doc
 5response = ollama.embeddings(
 6  prompt=prompt,
 7  model="mxbai-embed-large"
 8)
 9results = collection.query(
10  query_embeddings=[response["embedding"]],
11  n_results=1
12)
13data = results['documents'][0][0]

1# generate a response combining the prompt and data we retrieved in step 2
2output = ollama.generate(
3  model="phi3",
4  prompt=f"Using this data: {data}. Respond to this prompt: {prompt}"
5)
6
7print(output['response'])