The 2025 AI Coding Landscape - Coding models, agents, CLIs, IDEs, AI app builders, devtools, and more.
A work in progress... discovering the most recent developments in AI coding.
NOTES
The README.md file is regularly updated with new entries, the datasets and CSV files are updated bimonthly.
TO ADD OR UPDATE A LISTING
Want to add or update an AI coding model, agent, IDE, app builder, or tool in this README.md file? Let me know.
✅ Pull requests accepted!
OR
💬 Tweet at me @joy OR Connect with me on LinkedIn
OR
📩 Submit via email at jklarkin+AIMM AT gmail dot com
⤷ Are you a VC interested in working together? Let me know.
🙌 Spread the word about this repo.
🌟 GitHub stars are appreciated.
🔍 Search the AI Coding Tools CSV file
🔍 Search the AI Coding Models CSV file
🤗 Explore the AI Coding Tools HF dataset
🤗 Explore the AI Coding Models HF dataset
️🔥 Use the MCP server
📖 Read the blog post
- OpenAI Codex - Cloud coding agent toolkit
- GitHub Copilot - Pair-programming assistant
- Claude Code - Anthropic terminal agent
- Gemini Code Assist - Google AI coding assistant
- Jules - Google Asynchronous Coding Agent
- Cognition - Devin - An autonomous AI software engineer that can write, run and test code
- Amazon Q Developer - AWS code-gen & refactor
- Cursor AI - Agent baked into Cursor IDE
- Goose - Model + agent API
- Amp - Sourcegraph coding agent (CLI / VS Code)
- Reflection AI - Asimov - Enterprise code research agent
- Conductor - Run a bunch of Claude Codes in parallel
- Scout - Calls itself the most curious coding and research agent
- Blackbox AI - New Autonomous AI Coding Agent
- Forge Code - An AI software engineering agent that runs in your terminal
- Factory - Delegate software development tasks to agents called Droids
- Replit Agent - Set up and create apps from scratch, works with any framework
- JetBrains Junie - Your smart coding agent
- Slate - A purpose built agent designed to work with you for long and hard coding tasks
- GitHub Copilot CLI - The power of GitHub Copilot coding agent directly to your terminal
- Codebuff - Works in your terminal to help you write and deeply understand your code
- Aider - Terminal pair-programming
- Continue - IDE extensions + CLI
- Cline - Autonomous IDE agent
- Roo Code - Cline fork, VS Code extension
- Kilo Code - AI coding agent for VS Code and JetBrains
- Gemini CLI - An open-source AI agent for Google Gemini
- OpenAI Codex CLI - Open‑source command‑line agent for OpenAI
- OpenHands - Multi-tool coding agent
- Qwen Code - A command-line AI workflow tool for Qwen3-Coder
- Ruler - Central AI agent rule registry
- OpenCode - OSS terminal assistant
- Vibe Kanban - Orchestrate multiple agents
- Charm - A charming terminal agent, your new coding bestie
- Goose - An open source, extensible AI agent that goes beyond code suggestions
- DeepCode - Transforms research papers and natural language into production-ready code
- Visual Studio Code
- IntelliJ IDEA / PyCharm / WebStorm
- Xcode
- Eclipse, NetBeans
- Atom - Atom community fork
- Blackbox IDE
- Cursor - AI-first VS Code fork
- Windsurf - Agentic IDE, advanced AI coding assistant for developers and enterprises
- Zed - High-performance Rust editor with AI chat
- Amp - VS Code Extension
- Trae - ByteDance AI IDE
- Augment Code - Developer AI platform that helps you understand code, debug issues, and ship faster
- Warp - An agentic development environment
- Kiro - [Waitlist] Helps you do your best work by bringing structure to AI coding with spec-driven development
- Bolt - Browser-based AI app builder
- Lovable - Chat-to-app builder
- Replit - Cloud IDE w/ Ghostwriter
- v0.dev - Vercel text-to-UI generator
- Mocha - YC-backed no-code app builder
- Nectry - Responsible vibe coding for the enterprise
- Reflex - From prompt to production, build and deploy Python apps
- Superblocks - Build secure internal apps with AI
- vybe - Build internal apps 10X faster
- Emergent - YC-backed, build ambitious apps with agentic vibe-coding
- orchids v2 - YC-backed, the worlds first AI Full Stack Engineer
- Same - YC-backed, build fullstack web apps by prompting
- Aura - Generate beautiful designs in seconds and export to HTML or Figma
- 21st.dev - Build products that reflect the team's own taste
- Base44 - Lets you build fully-functional apps in minutes with just your words
- VibeFlow - YC backed, transform your AI-generated frontend mockups into fully functional applications
- Blink - Turn any idea into a beautiful, working app in seconds
- a0 - YC backed, ship mobile apps to the App Store and Google Play with AI
- Anything - Create powerful apps & websites by chatting with AI
- Rocket - Think It. Type It. Launch It.
- Rork - Builds complete, cross-platform mobile apps using AI and React Native
- Vibecode - Create native apps in seconds with AI
- bitrig - Build apps for your phone, on your phone
- Spielwork - The Tiktok for vibecoded mini games!
- Gizmo - A new way to make playful, personal software—right from your phone
- Hivemind - The fastest & easiest way to chat & code with any AI in one app
- Bloom - YC backed, go from idea to native mobile app on your phone without writing a single line of code
- Vibe Code Go - YC backed, code from your phone, a mobile app for software engineers
- Dyad - A local, open-source AI app builder
- Open Lovable - Clone and recreate any website as a modern React app in seconds
- bolt.diy - Bolt.new OSS version, AI-powered full-stack web dev for NodeJS based apps, choose the LLM you use for each prompt
- app.build - An open-source AI agent that builds full-stack apps
- ToolJet - An open-source low-code framework to build and deploy internal tools
- Adorable - Another open source Lovable version
- Vercel - OSS Vibe Coding Platform
- Cloudflare VibeSDK - Run an entire vibe coding platform end-to-end, with just one click
- Ollama - Chat & build with open models
- LM Studio - Run gpt-oss, Qwen, Gemma, DeepSeek on your computer
- Open WebUI - Self-hosted AI platform designed to operate entirely offline
- SillyTavern - A locally installed UI for text, image, and voice LLMs
- Unsloth - An open-source framework for LLM fine-tuning and reinforcement learning
- n8n - Flexible AI workflow automation for technical teams
- Firecrawl - Turn websites into LLM-ready data
- Agents.md - A simple, open format for guiding coding agents, used by over 20k open-source projects
- Vercel AI Gateway - A gateway to access hundreds of models with zero markup on tokens (including BYOK)
- OpenRouter - A unified API providing access to hundreds of AI models through a single endpoint
- Fabric - An open-source modular system for solving specific problems using crowdsourced AI prompts that can be used anywhere
- Vibetunnel - VibeTunnel proxies your terminals right into the browser, so you can vibe-code anywhere
- SWE-Bench Pro (Commercial Dataset) - A new benchmark designed to provide a rigorous and realistic evaluation of AI agents for software engineering
- Aider - Aider polyglot coding leaderboard
- SWE-bench - SWE-bench evaluates LLM performance on real world software issues collected from GitHub
- SWE-rebench - A Continuously Evolving and Decontaminated Benchmark for Software Engineering LLMs
- OpenRouter - Model, Market Share, Use Case Categories, and App Rankings
- Terminal-Bench - A benchmark measuring the capabilities of AI agents in a terminal environment
- PR Arena - Software engineering agents head to head
- Multi-SWE-bench - A Multilingual Benchmark for Issue Resolving
- SWE-DEV - Evaluating and Training Autonomous Feature-Driven Software Development
- LiveCodeBench - Holistic and Contamination Free Evaluation of Large Language Models for Code
Noteworthy releases, some entries may be updated model versions or model families.
- GLM-4.6 - Z.ai, features a longer context window, superior coding performance, advanced reasoning, more capable agents, and refined writing versus GLM-4.5
- Claude Sonnet 4.5 - Anthropic, the strongest model for building complex agents, the best model at using computers, it shows substantial gains on tests of reasoning and math
- Qwen3-Max-Instruct - Alibaba Cloud, the official release further elevates its capabilities — particularly in coding and agent performance
- GPT‑5-Codex - OpenAI, a version of GPT‑5 further optimized for agentic coding in Codex and trained with a focus on real-world software engineering work
- Kimi K2-Instruct-0905 - Moonshot AI, updated SOTA model with improved agentic and frontend capabilities and increased context length
- GPT-5 - OpenAI, flagship model
- GPT-5-mini - OpenAI, fast/cost efficient
- GPT-5-nano - OpenAI, faster/cost efficient
- Claude Opus 4.1 - Anthropic, a drop-in replacement for Opus 4
- Mistral Medium 3.1 - Mistral AI, aka Mistral-Medium-2508 - enterprise-grade model excels in coding tasks
- Grok Code Fast 1 - xAI, a speedy and economical reasoning model that excels at agentic coding
- Qwen3-Coder - Alibaba Cloud, agentic code model
- Qwen3-Coder-Flash - Alibaba Cloud, streamlined non thinking agentic code model
- Kimi K2 - Moonshot AI, 1 T-param MoE
- GLM-4.5 - Z.ai, an open-source LLM designed for intelligent agents
- Codestral 25.08 - Mistral AI, code model for high-precision fill-in-the-middle (FIM) completion
- Devstral Medium 2507 - Mistral × All Hands AI, high-quality and cost-effective model
- Devstral Small 1.1 2507 - Mistral × All Hands AI, agentic model
- Grok 4 - xAI, Grok 4
- Gemini 2.5 Pro - Google DeepMind, flagship model 5
- Gemini 2.5 Flash - Google DeepMind, fast/cost efficient with thinking capabilities
- Claude Opus 4 - Anthropic, pushes the frontier in coding, agentic search, and creative writing
- Claude Sonnet 4 - Anthropic, improves on Claude Sonnet 3.7 across a variety of areas, especially coding
- DeepSeek-R1-0528 - DeepSeek, OSS reasoning model
- o3 - OpenAI, preview reasoning model
- o4-mini - OpenAI, compact model
- GPT-4.1 - OpenAI, flagship model with 1M token context window
- Llama 4 Maverick - Meta, code-tuned model
- Llama 4 Scout - Meta, open-weight model
- Mellum - JetBrains, 4-B param OSS model
- DeepSeek-V3-0324 - DeepSeek, improved V3 version
- Gemini 2.0 Flash - Google DeepMind, multimodal for high-volume high-frequency tasks
- Claude 3.7 Sonnet - Anthropic, first hybrid reasoning model and state-of-the art for coding
- Grok 3 - xAI, Grok 3
*** SOURCE: joylarkin/AI-Coding-Landscape - https://github.com/joylarkin/AI-Coding-Landscape