Continue Coding Assistant
Overview
Continue is a coding assistant that lives inside of a VS Code extension. It supports chatting with your codebase, making edits, and a lot more.
Demo Video
Expectations
We have found that the Qwen-1.5-7B-Chat-Hybrid
model is the best Hybrid model available for coding. It is good at chatting with a few files at a time in your codebase to learn more about them. It can also make simple code editing suggestions pertaining to a few lines of code at a time.
However, we do not recommend using this model for analyzing large codebases at once or making large or complex file edits.
Setup
Prerequisites
- Install Lemonade Server by following the Lemonade Server Instructions and using the installer .exe.
Install Continue
Note: they provide their own instructions here
- Open the Extensions tab in VS Code Activity Bar.
- Search "Continue - Codestral, Claude, and more" in the Extensions Marketplace search bar.
- Select the Continue extension and click install.
This will add a Continue tab to your VS Code Activity Bar.
Add Lemonade Server to Continue
Note: The following instructions are based on instructions from Continue found here
- Open the Continue tab in your VS Code Activity Bar.
- Click the chat box. Some buttons will appear at the bottom of the box, including
Select model
. - Click
Select model
, then+ Add Chat model
to open the new model dialog box. - Click the
config file
link at the very bottom of the dialog to openconfig.yaml
. - Replace the "models" key in the
config.yaml
with the following and save:
models:
- name: Lemonade
provider: openai
model: Qwen-1.5-7B-Chat-Hybrid
apiBase: http://localhost:8000/api/v1
apiKey: none
- Close the dialog box.
- Click the chat box again. You should see
Lemonade
where you used to seeSelect model
. Ready!
Usage
Note: see the Continue user guide to learn about all of their features.
Here are some examples for trying out Continue. These examples assume you have cloned this repo and allowed Continue to index it.
Chat with Files
Open the Continue tab in your VS Code Activity Bar, and in the "Ask anything" box, type a question about your code. Use the @
symbol to specify a file or too.
- "What's the fastest way to install Lemonade in
@getting_started.md?
" - "According to
@README.md
what do I need to do to set up for@api_oga_hybrid_streaming.py
?"
Editing Files
Open a file, select some code, and push Ctrl+I to start a chat about editing that code.
- Open
//examples//lemonade//api_basic.py
. - Select the
print(...
line at the bottom and pressctrl+i
. - Write "Add a helpful comment" in the chat box and press enter.
- Press "accept" if you would like to accept the change.
Making Files
Start a new chat and prompt:
write a script in the style of
@api_basic.py
that uses the microsoft/Phi-4-mini-instruct model on GPU
Here's what we got:
# Import necessary modules
from lemonade.api import from_pretrained
# Load the Phi-4-mini-instruct model with the hf-cpu recipe
model, tokenizer = from_pretrained("microsoft/Phi-4-mini-instruct", recipe="hf-cpu")
# Define your prompt
prompt = "This is a sample prompt for the Phi-4-mini-instruct model"
# Tokenize the prompt
input_ids = tokenizer(prompt, return_tensors="pt")
# Generate the response using the model
response = model.generate(input_ids, max_new_tokens=100) # Adjust the max_new_tokens as needed
# Decode the generated response
generated_text = tokenizer.decode(response[0])
# Print the response
print(generated_text)