Comments (7)
@vicendominguez do you see any errors in the console for the Chrome extension's background script? Based on the server code, it looks like one of the validation steps is failing. Validation:
- request body is set
- model is set
- model is valid
- working directory is set
- embedding option is enabled
from lumos.
@andrewnguonly i found the problem.
TLDR
You need (must) to have llama2 model installed: ollama pull llama2
.
CONTEXT
llama2
model was not installed in my ollama serve
because my default LLM is mistral
.
I was trying to play with this line but the Llama2
model appears to be more integrated within the LangChain node module.
from lumos.
I'll update the README
to specify that llama2
is required because of the hardcoded model. Will also include instructions to change the model.
from lumos.
Not for me @andrewnguonly .....
diff --git a/src/scripts/background.ts b/src/scripts/background.ts
index 4d69d32..15cfb7b 100644
--- a/src/scripts/background.ts
+++ b/src/scripts/background.ts
@@ -9,7 +9,7 @@ import { formatDocumentsAsString } from "langchain/util/document";
const OLLAMA_BASE_URL = "http://localhost:11434";
-const OLLAMA_MODEL = "llama2";
+const OLLAMA_MODEL = "mistral";
var context = "";
chrome.runtime.onMessage.addListener(async function (request) {
Rebuilt, reloaded and inspected from your plugin (network tab). This is the request payload:
{"prompt":"texttextext","model":"llama2"}
That model llama2
has a bad smell....
Two questions:
- Could the
npm build
cache anything i don't know? (nodejs is out of my scope sorry) - Could you try it again but making a
ollama rm llama2
in your environment?
PS: i've updated nodejs to the last versions just in case.
from lumos.
I was trying to play with this line but the Llama2 model appears to be more integrated within the LangChain node module.
@vicendominguez, I was able to run the extension w/ mistral
after downloading the model and changing the hardcoded model value. Were you seeing some other error?
from lumos.
Could the npm build cache anything i don't know? (nodejs is out of my scope sorry)
I don't think so, but try deleting the entire /dist
folder and then rerun npm run build
to rebuild all the source files.
Could you try it again but making a ollama rm llama2 in your environment?
I tried this and the extension was able to call the API w/ the mistral
model. To clarify, I tried w/ both llama2
and mistral
downloaded and with only mistral
downloaded.
from lumos.
@vicendominguez, please reopen this issue if you're still unable to get the extension to call the Ollama API w/ the desired model. Thanks.
from lumos.
Related Issues (20)
- Lumos with WSL2 HOT 8
- Dark mode HOT 2
- Chat with PDF file in browser? HOT 4
- Skip download of unsupported image formats
- Enable chat model for images (multimodal support) HOT 1
- Sort imports, add functionality to linting
- Add dark mode option to README
- TypeError: Cannot read properties of undefined (reading 'includes') HOT 2
- Support cancel streaming
- Unable to connect to Ollama API on Mac OS HOT 7
- Support uploading image file (multimodal)
- Add shortcut to unattach file
- Add functionality to delete individual message (or regenerate LLM response)
- Add LangChain `YoutubeLoader` HOT 4
- Add audio document loader HOT 1
- WebLLM HOT 2
- Window resolution abnormal HOT 12
- Add support for embedding model `mxbai-embed-large`
- Summarize chat for chat title/preview (chat history view)
- Support `snowflake-arctic-embed` embedding model
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from lumos.