GPT-4 Copilot X is coming! AI code writing efficiency increases by 10 times, coders are hit by dimensionality reduction

Microsoft is really crazy!

Last week, Microsoft just upgraded the Office family bucket with GPT-4. Before people could react, Microsoft came to copy coders' homes again.

Today, GitHub officially announced: Copilot X, a new generation of code generation tool based on GPT-4, is here!

In less than two years since its release, GitHub Copilot has already written 46% of the code for more than 1 million developers and increased 55 % encoding speed. And this major upgrade is even more powerful.

The new generation Copilot Customized smart document systems and more.

GPT-4 supports Copilot, you can write code with just your mouth

Specifically, the experiences supported by Copilot X include: Copilot Chat; Copilot for Pull Request; Copilot for Docs; Copilot for CLI.

These new features are driven by OpenAI’s GPT-4. It is worth noting that due to speed delays, the code auto-completion tool is still based on GitHub's Codex model, which is trained on GPT-3.

Now, Microsoft has integrated GPT-4 into various real productivity tools such as search, office, and coding. It truly has the potential to start the fourth technological revolution. strength.

Senior big data architect Zhu William said that Everything powered by AI is no longer far away:

If , OpenAI GPT-4 is just a model that has completed its transformation from 0 to 1, and Microsoft has taken a big step forward in its commercialization.

This time, the release of Copilot X directly reduces the dimensionality to attack the previous generation Copilot.

Currently, Copilot X has opened the queuing entrance. But the more troublesome thing is that you have to apply to queue for each experience.

Netizens joked about this: "Every day I am on the way to apply for Waitinglist."

​——Just joined Bard yesterday The waiting list for Copilot X starts again today.

Copilot Chat

Application address: https://github.com/github-copilot/chat_waitlist_signup/join

GitHub embeds a GPT-4-based chat window in Copilot, focusing on developer scenarios, and integrated into VS Code and Visual Studio.

However, it is more than just a chat window.

Now, Copilot can not only identify the code content entered by developers and display error messages, but also conduct in-depth analysis and explanation of the purpose of code blocks and generate unit tests.

You can even give debugging suggestions.

Additionally, in Copilot you no longer even need a keyboard to write code.

Just sit in front of the computer, say "Hey, GitHub!", and write code with just a few words.

Currently, GitHub is experimenting with the new feature of GitHub Copilot Voice, a voice-based interaction system.

In addition to writing code, through Copilot Voice, you can even complete: code jump, control IDE, and code summary.

Copilot for Pull Request

Application address: https://copilot4prs.githubnext.com/login

In addition to the editor, the new GitHub Copilot will also provide support for Pull Requests.

Developers insert tags in the PR description, and Copilot will automatically recognize it and expand it.

Currently supported tags mainly include the following:

• copilot:summary generates a summary summary for the PR.
• copilot:walkthrough a detailed list of changes, including links to relevant code snippets.
• copilot:poemWrite a poem to describe this change.
• copilot:all automatically generates all of the above.

And this is only the first step.

In the future, Copilot will also add two new functions: Gentest and Ghost Text.

Gentest: When Copilot finds that a Pull Request submitted lacks sufficient tests, it will automatically remind the developer and suggest potential tests that can be edited, accepted, or rejected based on project needs.

For project owners, this feature can also help them formulate strategies for testing and support developers in following these strategies.

Ghost Text: When a developer creates a Pull Request, Copilot will dynamically obtain relevant information about code changes and automatically give suggestions.

In addition, other ongoing explorations and experiments on GitHub include:

Solving AI problems: Yes Automatically describe how to solve the problem and even automatically suggest changes that need to be made.

As follows, after submitting an issue transferred to PyTorch, the new function can provide generated code suggestions and initiate a new Pull Request.

Use AI to review PRs: Since a large number of developers spend time on code reviews and preparing PRs for code reviews, this feature can reduce the review cycle.

Copilot for Docs

Application address: https://githubnext.com/projects/copilot-for-docs

The importance of an excellent technical document to a developer is obviously self-evident.

Copilot for Docs This function can help you quickly locate and find the content you want from this document.

The official blog states that Copilot for Docs is only an experimental tool. Programmers can use the chat interface to find content in the documentation, including issues such as languages, frameworks and technologies used.

The most amazing thing is this personalized feature.

It can output different answers based on the user's different programming levels, understanding of the document, and what they want to find.

Programmers can also get the latest content as soon as possible, keeping existing materials in sync with the latest version of documentation directly from the GitHub repo.

At the same time, another benefit of getting content from GitHub is that questions about the team's private content can be easily answered. This feature can provide information that only your internal team can see based on existing GitHub permissions.

In addition, Copilot can also shuttle back and forth between different third-party documents, such as React, Azure documents, and MDN, and can splice their contents and return the results.

In the future, GitHub will also explore other ways of indexing resources beyond documentation, such as issues, requests, discussions, and wikis, to provide developers with everything they need to answer technical questions.

Copilot for CLI

Application address: https://githubnext.com/projects/copilot-cli/

Next to code editing and pull requests, the command line terminal is where developers spend the most time.

But even the most skilled developers need to scroll through many pages to remember the precise syntax of many commands.

And the GitHub Copilot CLI can write commands and loops and give obscure find flags to satisfy the programmer's query.

CLI's Copilot provides three shell commands: ??, git?, gh?.

"??" refers to the general goto used for any shell command. It will compose commands and loops and throw out obscure search flags to satisfy individual queries.

"git?" is used to specifically search and call git. Compared to ??, it will be more powerful in generating Git commands, you don't even need to explain it in the context of Git, and the query will be more concise.

"gh?" will combine the power of GitHub CLI commands and query interfaces, and use AI to generate complex flags and jq expressions.

The golden age of software development! 10x efficiency improvement

GitHub CEO Thomas Dohmke said that although automatic code completion has greatly improved developer productivity, the new Copilot X can increase developer productivity 10 times.

With the advent of new chat features, Copilot will become the foundation for people to learn to code in the future.

"The golden age of software development has arrived in the chat interface."

NVIDIA scientist Jim Fan Said, "10x engineers are still a myth, but 100x engineers powered by AI are more real than ever."

In view of GitHub "Mixed" OpenAI's GPT-3 and GPT-4 to drive the chat function and the existing auto-completion code function.

Dohmke explained that when you type in an editor, you definitely want to get a very fast response.

Therefore, where speed is needed, we use smaller models, such as code auto-completion based on Codex; and where accuracy is needed, we use larger models, Such as using GPT-4 in chat.

Netizens: Coders are "dead"

Netizens have said that programming has been forever changed by ChatGPT.

"As a developer, this is really frustrating. In programming, our source of joy is working with logic and solving technical problems, rather than focusing on productivity or meeting needs."

"I've been programming for 30 years and I still love it. I don't know what a world would be like without human code. But I do love having CoPilot to handle the annoying stuff. , and ChatGPT to help me solve bugs."

"After the emergence of AI, there may be no need for so many coders. In this regard I feel very anxious. Even though I have saved money and invested in various ways, it will still take almost 9 years to "sit down". I am slightly envious of those who work in FANG, they seem to be able to achieve financial freedom easily."

"In the medium to long term, we will lose our jobs."

Programmers are finally using Programming eliminates itself...

The above is the detailed content of GPT-4 Copilot X is coming! AI code writing efficiency increases by 10 times, coders are hit by dimensionality reduction. For more information, please follow other related articles on the PHP Chinese website!