You have 1 article left to read this month before you need to register a free LeadDev.com account.
Estimated reading time: 8 minutes
There has been an explosion of generative AI programming tools hitting the market over the past three years, many of which are free to use.
These tools have quickly shown that they can fit neatly into software developers’ workflows and provide an instant productivity boost. They’ve very quickly gone from interesting novelty to essential tool.
These tools and agents – which leverage large language models (LLMs) and other kinds of AI models to understand language, imagery, and code – can modify, test, review, and ship code across whole repositories, as well as explain what’s going on in a codebase.
While they’re improving at a rapid rate, these tools aren’t perfect yet, and still require oversight from experienced engineers to get the most from them.
As with any tech boom, there are suddenly dozens of different programming tools vying for engineers’ attention. For 2026, we revisited our list and added any new tools that were getting traction, removed a few old ones that had stalled out, and otherwise made sure that we only featured the best tools you should really know about.
Your inbox, upgraded.
Receive weekly engineering insights to level up your leadership approach.
Claude Code
Anthropic’s Claude is ahead of the pack when working directly with a codebase. You can use Claude Code through any Claude app or integrate it with your local dev environment and use it through the terminal or your integrated development environments (IDE). It can then explore your codebase or work directly on it.
“We’ve had good results using Claude Code for tasks that involve understanding intent, rather than just generating syntax,” says Krunal Vyas, CEO of development studio iQlance. “Developers often use it to review long functions, refactor messy legacy code, or sanity-check business logic. It’s less about speed and more about clarity, which makes it useful during code reviews or architectural cleanups.”
Claude is free to try, though you need the $20/month Claude Pro plan if you want full access to Claude Code. There are usage caps so if you come to rely on it, you will have to pay upwards of $100/month for Claude Max.
ChatGPT
OpenAI’s ChatGPT is the free app that made the world aware of what generative AI could do. While the chatbot might not be the best app for programming tasks specifically, it is still able to generate boilerplate code, translate or refactor lines of code into different languages, automate rote tasks, and even help with more complex coding problems.
While you don’t have to pay, install anything, or give it access to your codebase to see if the suggestions are good, or if it can explain what complex functions do, ChatGPT Plus at $20 per month gives you more access to OpenAI’s most powerful models. This makes ChatGPT a good starting point when it comes to exploring what a generative AI tool can do. While it can provide good solutions to some problems, there are better coding-focused tools out there.
OpenAI Codex
Codex is OpenAI’s answer to Claude Code. It’s a cloud-based agent that can work directly with the code in a GitHub repository. You can control it from the ChatGPT app, or integrate it with your IDE or terminal. It uses a fine-tuned version of OpenAI’s GPT-5 reasoning model. You’re able to both ask it for information about your codebase and direct it to make changes, like fixing bugs and adding new features.
In a detailed review for Infoworld, Martin Heller calls it a “first-rate coding agent” and says it “rivals Claude Code.”
Codex runs everything in a sandbox without internet access, so it can test any code it creates in a safe environment before submitting a pull request. It’s available to ChatGPT Plus ($20/month) and ChatGPT Pro ($200/month) subscribers.
More like this
GitHub Copilot
GitHub Copilot is meant to be an AI pair programmer. Unlike chatbots, it’s explicitly designed to be capable of writing functional code in languages like Python, Javascript, Go, PHP, Ruby, and Swift.
Because Copilot has full access to your project, it is able to generate specific code suggestions based on architecture, style conventions, and context. If you ask it to create a function it won’t just give you boilerplate code – it will do it using the variables, classes, and methods you’ve already defined. You can also select what model from OpenAI, Google, or Anthropic you want it to use to generate your code suggestions.
“In day-to-day development, GitHub Copilot is still the tool most teams reach for first,” says Vyas. “It’s especially effective for speeding up routine work like scaffolding components, writing tests, and filling in predictable patterns. Where Copilot struggles a bit is with larger context, developers still need to guide it carefully once logic becomes complex.”
Copilot integrates with a small number of popular IDEs, including Neovim, Visual Studio, and Xcode.
Cursor
While Copilot, Codex, and Claude Code are add-ons available in multiple IDEs, Cursor is an AI-first IDE itself. Instead of AI being a useful if optional extra, with Cursor it’s a core part of the coding experience.
Features like Composer allow the AI to work on multiple files at once in an autonomous, agent-like manner, and Cursor’s access to your whole code-base means it can make intelligent suggestions based on your existing functions.
Although it’s only been available since 2023, it’s already proving to be incredibly popular with engineers and was the most popular tool in our AI Impact report last year.
“For serious work, I’m all in on Cursor,” says Arshad Yaseen, founding engineer at Invent. “Their tab completion is honestly insane – it’s like it’s reading your mind. You just tab tab tab and entire blocks of code appear exactly how you’d write them. The speed is unmatched, and their agents are solid too. I’ve tried others, but nothing comes close to Cursor’s tab feature yet.”
Google Gemini and Antigravity
Gemini is Google’s response to ChatGPT. Google developed the transformer architecture that has underpinned the recent boom in LLMs, so it has a huge amount of AI expertise, but it was slow to leverage that into usable products.
Google Antigravity was announced late last year and is currently available in public preview. It’s an AI-first IDE and allows you to use the latest Gemini model as well as Claude and OpenAI’s open GPT-OSS models. Right now, it’s free to use with generous rate limits, but that is likely to change as it becomes a more mature part of Google’s product line up.
Yaseen calls it “big news” and says “it’s getting hyped.”
Zed
Like Cursor, Zed is a modern IDE designed to enable collaboration between humans and AI. It’s open-source and you can use it for free with your own API keys for AI, or plans start at $10/month if you want support and enterprise features.
Zed stands out for its speed and ability to efficiently use multiple CPU cores as well as your GPU. It also includes collaborative chat and notes features.
Yaseen says he uses it a lot because it’s “stupid fast… and the whole thing feels snappier than anything else.”
Tabnine
Tabnine is another AI coding assistant that supports a range of LLMs from OpenAI, Anthropic, Mistral, and more to generate code suggestions. It can auto-complete whole lines of code or full functions, as well as match the style and conventions of your existing codebase.
Where Tabnine stands out is with its availability. Most of the tools on this list are only available in a handful of major IDEs, like VS Code, whereas Tabnine is also available in a wider variety of apps, including Sublime, Eclipse, Neovim, and Android Studio.
On Gartner’s Peer Insights, one recent review calls it “a boon for our engineering organization” and highlights its privacy focus.
OpenCode
OpenCode is an open source alternative to tools like Claude Code and OpenAI’s Codex. It also allows you to use the AI coding agent through your terminal, IDE, or a dedicated app, but it gives you a lot more freedom and flexibility. You can connect any model you want and even run models locally.
OpenCode is quietly popular. It has more than 500 contributors on GitHub and is used by over 650,000 developers every month.
Yaseen says he’s been using OpenCode over Claude Code because its UI feels faster and he likes the custom themes. He also says that “Claude Code asks permission for every little thing which gets annoying, while OpenCode intelligently decides what needs permission.”
Amp (formerly Cody)
Amp from Sourcegraph bills itself as a “frontier coding agent”. It combines multiple models, including Claude Opus, Gemini 3 Pro, and GPT-5.1, and intelligently selects which one is most appropriate for each task. For example, Claude Opus is used as a coding agent while GPT-5.1 is used for complex reasoning and making plans on how to tackle problems. It integrates with both the terminal and IDEs like VS Code, Cursor, and Neovim.
Amp offers $10 of free credits per day. After that, it uses a credit system. Different models require different amounts of credits, so it really does vary based on how you use Amp.
A user review on G2 says Amp “stands out for how quickly and effortlessly it allows our team to deploy and manage content or processes. Its clean interface and logical layout mean even new users can become productive without a steep learning curve.” It also highlights how it integrates with existing tools as well as the responsive support team.
ChatGPT Image Generation, FLUX.1, and Nano Banana
It’s not just back-end developers who can get the benefits of the current generative AI boom. Text-to-image tools like ChatGPT Image Generation, FLUX.1, and Nano Banana (yes, that’s really what it’s called) allow you to generate any image you want from a simple, written prompt to quickly generate app icons, interface elements, and placeholders for any front-end design elements.
If you don’t have the time – or the design chops – and need something workable right now, even just as a placeholder, they’re surprisingly effective.
