Introducing Code Index Graph for Seamless AI-Assisted Codebase Management
We are excited to announce the limited release of our new Code Index Graph Tools, designed to enhance your workflow by integrating AI assistance with your full codebase repository
How to Get Access:
✨
Llama 3.1 Now Available in VSCode as a Code Assistant via
@GroqInc
🚨
You can now use this new
@AIatMeta
model directly in VSCode using the
@codegptAI
extension
My first impression: The model is incredible 🚀
NVIDIA CodeLlama 70b as a Copilot in VSCode 🤩
Let me show you how to connect the NVIDIA CodeLlama 70b model in VSCode
Here is a step-by-step tutorial! 🧵👇
Introducing Ollama fully Integrated into VSCode with CodeGPT
Now you can download open-source models from
@ollama
with a single click and install them as code assistants in VSCode.
In this case, I download the qwen2:0.5b (
@Alibaba_Qwen
) model in under a minute, and it becomes
Introducing CodeGPT, running the
@MistralAI
7B model locally in VSCode on an M2 MacBook Pro 🤯
In my trials, Mistral 7B surpasses Llama 2 and CodeLlama in both speed and performance.
Feel like giving a try?
You can try it for free! Just download the CodeGPT extension from the
Use Copilot for free and completely privately with the DeepSeek models in VSCode
Follow these 5 steps:
- Download the CodeGPT extension for VSCode:
- Download
@Ollama_ai
:
- Once installed, run the following command in the
Introducing
@MistralAI
as your copilot in VSCode! 🚀
Effortlessly integrate Mistral AI with CodeGPT to enhance your coding experience.
Try out the models:
- mistral-tiny
- mistral-small
- mistral-medium
Check the documentation here:
Created with
Claude 3.5 Sonnet is now available as a code assistant in VSCode!🚨
If you already have the extension installed, you just have to reload VSCode and Claude 3 will automatically update to Claude 3.5 🙌
In the video I show how to document a code in python and it also adds the
Copilot is now completely free and private thanks to
@LMStudioAI
and the OpenHermes-2.5-Mistral-7B model
Finally,
#LMStudio
is seamlessly integrated with CodeGPT, allowing you to leverage all models of this incredible platform locally in VSCode.
Download CodeGPT extension for
Llama3 70B as a copilot in VSCode thanks to the integration with Perplexity 🤩
The opensource llama3 70b model released by Meta has already surpassed many private models.
Now you can use it by connecting to
@perplexity_ai
through its API.
In this thread 🧵 I tell you how to
Free and private Copilot in the middle of a flight and without internet access with
@MistralAI
😨
Instructions:
- Download and install
@Ollama_ai
here:
- Download
@codegptAI
here:
- Run this commands:
“ollama pull mistral”
Llama3-70b and phi-3-128k as Copilot in VSCode powered by NVIDIA AI 🤯
Now you can use these two models within VSCode using the
@NVIDIAAI
API
In this thread 🧵, I'll walk you through the integration process, explaining each step simply and clearly👇
Your own Copilot in VSCode lightning-fast thanks to Groq and Llama3
In this thread 🧵, I'll guide you through connecting
@GroqInc
with VSCode, unlocking the full potential of Llama 3 (
@AIatMeta
) as your Copilot👇
Llama3 as Copilot in VSCode completely private and free! 🤩
In this thread 🧵, I share with you two step-by-step tutorials to use Llama3 running locally on your computer with
@ollama
and
@codegptAI
Enjoy VSCode Copilot for free! 👇
Claude 3 as a Copilot in VSCode is a game changer.
Every day, I program more and more using Claude 3, and it keeps getting better.
Thanks
@alexalbert__
and the Anthropic team, for creating such an incredible model 👏👏👏
In this thread🧵 I'll explain the different ways you can
Introducing Llama3 as Copilot in VSCode, now running locally on GPU using
@NVIDIAAI
TensorRT and Triton Inference Server 🤩
Running your own Copilot locally and privately is fantastic, but what's even more remarkable is being able to run it on an H100 GPU.
This integration is
Gemini-1.5-flash as a Copilot in VSCode is amazing!
You can now use this model by connecting CodeGPT with Google AI Studio.
@codegptAI
+
@googleaistudio
In this video, I show how CodeGPT manages to get the entire context of the "Quick Fix" section and Gemini provides a
Finally! Codellama-70b as Copilot in VSCode! 😱
You can now harness the power of the most advanced code generation model, Code Llama 70B, right in Visual Studio Code with
@perplexity_ai
This model has outperformed even GPT-4 in code writing, and it's now seamlessly integrated
Groq in VSCode: Explaining and Documenting Entire Files in Seconds! 🤩
Llama3 70B by
@AIatMeta
running on
@GroqInc
and using
@codegptAI
extension is an excellent combination of tools for programming in VSCode and JetBrains
Did you know that now you can interact with entire
This isn't Copilot 😱!
This is
@codegptAI
with your own llama, codellama or mistral model running locally with absolute privacy for your code
Available from version 2.1.28 onwards 🤩
Powered by
@LangChainAI
and
@Ollama_ai
Download the extension here:
Fully local Code Assistant running on NVIDIA GPU!
In this tutorial, I'll show you how to run Llama3 using TensorRT and Nvidia's Triton Inference Server to use it as a Code Assistant in VSCode
In this thread 🧵, I'll walk you through the integration process, explaining each step
GPT-4o-mini is now available for use as a Code Assistant in VSCode with CodeGPT 🚨
I highly recommend switching to GPT-4o-mini instead of GPT-3.5-turbo
The model is more cost-effective, faster, with a 128k context window, and the responses are significantly better
We did it! We’ve finally integrated Groq with VSCode as a Copilot 👏
This means that the response when sending a large amount of code is incredibly fast! 🚀
Mixtral 7b and Llama 2 70b models are available.
Follow these steps to use it:
✨ Download the CodeGPT extension at this
Introducing Code Assistant that Runs Code in VSCode!
We've enabled code execution for Python 3.10 Agent in a sandbox environment, thanks to the integration with
@e2b_dev
🙌
Debug your code before integrating it into your project with just a click of "Run Code" 🤩
Currently
How to use Microsoft's phi-2 as a copilot in VSCode and ask for code explanation to run phi-2 in Python🤯
In this example, I'm using Ollama to download the model, and then I just need to type the downloaded model's name into CodeGPT to use it as Copilot in VSCode.
The most
The future of app development is Agentic 🤖 just as
@AndrewYNg
explained
Now you can select different LLMs and different agents to work with them in VSCode and Jetbrains
Download the FREE CodeGPT extension here:
IBM + NVIDIA as your code assistant in VSCode 🤩
@IBM
has released its Granite Code Model family!
You can now use it in VSCode with the CodeGPT extension
To get started, simply select
@NVIDIAAI
as the provider and choose the granite-34b-code-instruct model.
Enjoy this
Introducing Google Gemini Pro as Copilot in VSCode 🚨
A new feature I've added to CodeGPT
It's really making a difference in the coding process.
I ask Gemini to help me with the Stripe SDK in PHP
More details in the documentation:
#CodeGPT
#GeminiPro
Code autocompletion with
@AnthropicAI
Claude 2 🚨
You can now select Claude 2 to autocomplete your code in VSCode 🤩
CodeGPT Copilot available since version 2.2.1
Install:
Docs:
It is integrated with LangSmith to track all the
Gemma2 vs Llama3 running simultaneously in VSCode thanks to
@ollama
and
@codegptAI
🤩
Ollama can now handle multiple requests simultaneously, allowing development teams to run different models on a single server for various use cases and connect them to VSCode via CodeGPT
Introducing Codestral + NVIDIA as a Code Assistant in VSCode! 🚨
You can now use Codestral by
@MistralAI
through
@NVIDIAAI
services via the CodeGPT extension in VSCode 🙌
If you want to try it, follow these steps 👇
In the video, I show how Codestral
Big News! 😱
CodeGPT is already integrated with
@huggingface
Available models:
google/flan-t5-xxl
HuggingFaceH4/starchat-beta
tiiuae/falcon-7b-instruct
Documentation:
@judiniAI
@codegptAI
Introducing
@StackOverflow
searches with GPT-4o in VSCode
This new feature allows you to use the capabilities of /StackOverflow find related posts ensuring a far more accurate answer to your question.
There's no waiting list! 🤩
You can use this today! All you have to do is
Creating your own Copilot in VSCode is ridiculously easy 🤯
In this video, using the CodeGPT extension in VSCode, I select OpenAI and the gpt-4-1106-preview model to review Python 🐍 code.
You can download the extension for FREE at this link:
Writing code with
@cohere
command-r-plus is an absolute pleasure! 🤩
In this thread 🧵, I'll walk you through the step-by-step process of obtaining an API Key
Then connecting it in VSCode to harness this incredible model 👇👇👇
The new version of CodeGPT is already in pre-release in VSCode, Cursor and JetBrains! 💥
Stay tuned as I'll soon demonstrate the capabilities of AI Teams within your code editor 👀
First code interpreter tests in a CodeGPT Agent 😱
In a few days, you'll be able to run Python 🐍 code with any model directly within VSCode, right in the chat 🙌
@e2b_dev
+
@codegptAI
If you're working with LLM, it's mandatory that you read this article 😱
"Evaluating a RAG System: Part 1 of 3"
Article link:
This is the first research article released by the CodeGPT team.
In this initial part, we employ the RAGAS framework to
Groq speed as a Copilot in VSCode is a game changer! 💥
In this video, I use Llama3-70b through
@GroqInc
in VSCode with the
@codegptAI
extension to document code
It's that easy to use the shortcuts:
✨ Fix
✨ Explain
✨ Document
✨ Unit Test
Enjoy these features, it's
Code autocomplete in VSCode using Mistral's Codestral model is impressive 😍
You can now enable Codestral for chat and code autocomplete by adding your
@MistralAI
API Key
✨ CodeGPT allows you to configure the max token and the time delay for suggestions to appear in the
Try Tinyllama as a copilot in VSCode 🤩
Run "ollama pull tinyllama"
Then select the model in VSCode and start working with the code in the CodeGPT Chat
Download CodeGPT extension here:
@codegptAI
@OLLAMA
Exciting news! Codestral Mamba by
@MistralAI
is now available as a code assistant in VSCode! 🚀
You can start using this powerful model right away through the
@codegptAI
extension
Simply open the model selector, and you'll find the newly released Codestral Mamba ready to use.
Now you can use Claude 3 as a Copilot in VSCode! 🚀
In this article, I share three ways you can integrate the amazing Claude 3 model from Anthropic into VSCode thanks to the CodeGPT extension.
Full article:
Enjoy Haiku, Sonnet, and Opus as your
Testing Code Interpreter with any LLM in Visual Studio Code
@codegptAI
open a notebook, create the code, run, repeat!
Very soon in the next version 🚀
Created with
@LangChainAI
@hwchase17
@DataChaz
🚀 New Feature Alert: Code Execution and Self-Correction with CodeGPT
We're about to release an incredible feature: the ability to run code directly during chat interactions!
Imagine code assistants that can not only write but also execute and self-correct their code
Codestral is now integrated into VSCode with
@ollama
models 🤩
You can try it out in chat mode today, and we're already working on adding autocomplete integration.
Congratulations to
@sophiamyang
and the
@MistralAI
team 👏👏👏
The model is an excellent coding assistant.
Codestral as a code assistant in VSCode completely free!
You can now use this model created by
@MistralAI
for free with the CodeGPT extension.
✨ Download the extension:
✨ Open the CodeGPT Tab and select Codestral
In this video, I test it with
If you want to learn more about
@LangChainAI
Agents, stop reading and start coding!
I took all the agents and all the tools and created a website with
@streamlit
to test them🤩
Very soon we will organize a workshop in Spanish with
@Microsoft
about agents!
@hwchase17
@MSFTChile
CodeLlama + Code Interpreter with CodeGPT 🚀
You can now use llama 2 in codegpt thanks to the integration with
Installation:
Can you find the error in the code that codellama gave me?
@hwchase17
@DataChaz
@LangChainAI
@jmorgan
Rabbit R1 as Copilot in VSCode ??? 🤯
It's very easy to create your own Code Assistant in CodeGPT
You can select the model, customize it and add your own data
In this thread 🧵, I share the step by step tutorial on how you can create your own Copilot and connect it to VSCode!
Creating a CRUD with FastAPI using CodeLlama locally with and CodeGPT integration
Available models:
llama2 (7b,13b,70b)
codellama (7b, 34b, python7b-34b)
phind-codellama (7b, 34b)
wizardcoder(7b, python7b-34b)
Powered by
@LangChainAI
Introducing Commit Autocomplete in VSCode with Codestral by
@MistralAI
🚨
@codegptAI
extension now includes a commit autocomplete feature for your repositories
After adding your changes, you can start typing your commit message, Codestral model will then autocomplete the
I highly recommend trying
@cohere
Command-R model
I have had incredible results with this model in both English and Spanish.
Fast and accurate responses, a huge improvement over their previous Command models.
If you want to try it, you just need to create an account at
Introducing Gemma by
@Google
as Copilot in VSCode for FREE! 🤩
Using
@ollama
and
@codegptAI
, you can load the Gemma:2b (or 7b) model to assist with code in Visual Studio Code.
Run the following command: ollama pull gemma:2b
Then select Ollama as the provider and type gemma:2b
Create React components with GPT-4-vision within VSCode 😱
Simply take a screenshot of any graphical interface and the chat will generate the corresponding React component for seamless integration into VSCode.
Download CodeGPT extension: …
Sign up for a
A pdf invoice converted to sql, html and php🤯
With
#CodeGPT
it is possible to program code with a base that can be a pdf, video or even another github repo.
Waitlist for this function:
@DataChaz
@midudev
@MoureDev
@shota7180
CodeGPT is now available in PyCharm and all
@jetbrains
IDEs 🙌
If you use JetBrains IDEs, you can download the plugin from this link:
Or simply search for CodeGPT in each IDE's marketplace.
In which other editors or IDEs would you like to use CodeGPT?
Create your AI Agent with the comprehensive knowledge of a GitHub repository's codebase
For programmers, the landscape is continually shifting with the disruptive advancements in Generative AI.
One of the most potent capabilities this technology offers is the ability to
Evaluating a RAG System: Part 3 of 3
Here is the link to the third part of the article:
In this part, we focus on Reranker with
@cohere
models.
When it is really convenient to use Reranker depends on which chunking system you are using (
@LangChainAI
,
Turn your simple drawing into a complete cloud setup using CodeGPT and Terraform! 👏
Ever heard of Terraform? It's a powerful tool that helps you build a cloud infrastructure using code.
You can deploy your entire application with just a simple image! 🤩
The best part is that
You will be able to import libraries and frameworks into CodeGPT and write your code with the base and context
In this example we import the complete and updated code from
@LangChainAI
Waitlist for this function:
@DataChaz
@midudev
@MoureDev
@_nasch_
If you have any doubts about a script that is too big to copy to
#chatgpt
Go to
#FileGPT
, upload the file, and start chatting with the script👇
Powered by
@LangChainAI
Grok Vision vs CodeGPT Python Agent 😱
The same image and the same question, but CodeGPT is FREE and you can use it in VSCode
✨ Download the extension:
✨ Create an account:
What do you think is the best one?
Tomorrow, I will deliver a presentation at
@Google
Chile regarding the development of agents using
@LangChainAI
and
@judiniAI
.
During the talk, I will discuss topics such as
@GoogleAI
, MakerSuite, VertexAI, and PaLM 2 models
gpt-4-turbo is incredibly fast! 🤯
A few months ago I did this same exercise with gpt-3.5-turbo-16k and in this case, apart from giving me the code, it also gave me instructions on how to install and run FastAPI, even using pydantic 😱
If you want to try gpt-4-turbo
We are going to release our API next week 😱
- Create an account at
- Upload documents
- Create an expert agent on these documents
- Connect with LangSmith to control consumption
- Deploy your agent to production 🤝
Powered by
@LangChainAI
Here I am showing the repository of
@OpenAI
Python SDK ready to be processed by CodeGPT to be able to integrate it as a Copilot in VSCode.
This is how we are working with large code repositories to provide good context to the LLMs so they can assist us in writing code.
If you
Happy to announce that
#CodeGPT
already has more than 80,000 installations in its first month and with this incredible record it is now possible to use
@CohereAI
in the extension! 👏
Two models available:
-command-xlarge-nightly
-command-medium-nightly
Introducing AI Agent with full codebase via Slack 🤩
Now you can create an AI agent expert in your code, and not only use it in VSCode but also interact with it through
@SlackHQ
Create a free account on and try it yourself!
Free and Private Copilot: The Future of Coding with LLMs in VSCode
This article provides a step-by-step guide on how to use open-source LLMs (
@MistralAI
) with the CodeGPT extension in VSCode
It also discusses the advantages and disadvantages of running these models locally on
Copilot expert in TypeScript for private and free use!
Introducing the "deepseek-coder-1.3b-typescript" model.
This model is a fine-tuning of deepseek-coder-1.3b trained on a dataset of 0.5B lines of TypeScript code.
Download the CodeGPT extension:
We are happy to welcome
@nvidia
to CodeGPT’s model providers 🙌
Now you can use available models from NVIDIA within VSCode, like Copilot, to work with your code.
Available models:
✨ meta/codellama-70b
✨ meta/llama2-70b
✨ google/gemma-7b
In this video, I’ll show you how I’m
Gemini 1.5 Pro is now available in 180+ countries and is already accessible on CodeGPT for use within VSCode.
✨ Select Google AI Studio as your provider.
✨ Add your API Key.
✨ And select gemini-1.5-pro.
You can now use this Google model to write code! 🙌
@Google
@GoogleAI
What is the best model for programming? let me tell you 🧐
I decided to create the 'Login with Tailwind Test' so that you can judge for yourselves which model is better
I'll showcase each of the tests I conducted, using
@codegptAI
in VSCode to create a simple login web page