feat: 🎸 local

all local in 60 seconds

.env.example

@@ -6,6 +6,9 @@ CREATE_FIRST_USER=true # set to false if you want to create the first user manua
 # login: [email protected]
 # password: admin
 
+#LOCAL
+#OLLAMA_API_BASE_URL=http://host.docker.internal:11434 # Uncomment to activate ollama. This is the local url for the ollama api
+
 
 
 
@@ -39,9 +42,6 @@ GOOGLE_CLOUD_PROJECT=<ignore-me>
 CELERY_BROKER_URL=redis://redis:6379/0
 CELEBRY_BROKER_QUEUE_NAME=quivr-preview.fifo
 
-#LOCAL
-#OLLAMA_API_BASE_URL=http://host.docker.internal:11434 # local all in one remove comment to use local llm with Ollama 
-
 
 #RESEND
 RESEND_API_KEY=<change-me>

README.md

@@ -70,16 +70,18 @@ Ensure you have the following installed:
 
 
   > Want to use [Ollama.ai](https://ollama.ai) instead?
-  > Finish installing Quivr, put a randon open_api_key in the `.env` file and
-  > follow the instructions [here](https://brain.quivr.app/docs/Developers/selfHosted/run_fully_local) to update Quivr to use Ollama AFTER you have finished installing Quivr.
+  > Uncomment the following lines in the `.env` file:
+  > OLLAMA_API_BASE_URL
+  > Run the following command to start Ollama: `ollama run llama2`
+  > You can find more information about Ollama [here](https://ollama.ai/).
 
 - **Step 4**: Launch the project
 
   ```bash
   docker compose pull 
-  docker compose up --build
-  ```
-
+  docker compose up --build # if OPENAI 
+  # docker compose -f docker-compose-ollama.yml up --build  # Only if using Ollama. You need to run `ollama run llama2` first.                
+```
 
 - **Step 5**: Login to the app