// AI Coding Assistants DirectoryCursor · Claude Code · Copilot · Cline · Aider · Windsurf·Every major tool scored and compared·Editorial scores · Pricing · MCP support · IDE coverage·Updated monthly// AI Coding Assistants DirectoryCursor · Claude Code · Copilot · Cline · Aider · Windsurf·Every major tool scored and compared·Editorial scores · Pricing · MCP support · IDE coverage·Updated monthly
AI Coding Assistants Directory
AI CODING
ASSISTANTS
Every major AI coding assistant scored and reviewed. We evaluate IDE support, agentic capabilities, pricing, MCP compatibility, model flexibility, and real-world developer sentiment — so you can pick the right tool for your workflow.
27
Tools reviewed
Scored 1–5 · IDE flags · Pricing · MCP support
// Featured
Open Source
MCP5/5
Claude Code
Anthropic
Terminal-first AI coding agent with the highest SWE-bench score (80.9%), 1M token context, and the highest developer satisfaction rating in the market.
Best for: Senior developers and teams who prefer terminal workflows, need large codebase understanding, and value reasoning depth over UI polish.
CLIVS CodeJetBrainsNative AppWeb
Mode: CLI / Terminal·Pricing: Freemium
Proprietary
MCP5/5
Cursor
Anysphere
AI-native IDE built on VS Code with the best autocomplete, multi-model flexibility, and Background Agents that run tasks on cloud VMs in parallel.
Best for: Professional developers and teams who want the most polished AI-native IDE experience with background agents.
Native AppJetBrains
Mode: Agentic·Pricing: Freemium
Proprietary
MCP4/5
GitHub Copilot
Microsoft / GitHub
The enterprise standard AI coding assistant with the broadest IDE support, 4.7M paid subscribers, and deep GitHub ecosystem integration.
Best for: Enterprise teams and organisations already invested in the GitHub ecosystem.
VS CodeJetBrainsNeovimCLIWeb
Mode: Agentic·Pricing: Freemium
Proprietary
MCP3/5
Devin
Cognition AI
The pioneer of autonomous AI software engineering. Works independently in sandboxed environments to turn task descriptions into pull requests.
Best for: Teams with well-defined, repeatable tasks where fully autonomous execution is viable.
Web
Mode: Autonomous·Pricing: Paid
// All Tools
Open Source
4/5
Aider
Paul Gauthier (community)
The original terminal AI pair programmer. Git-native by design, every AI edit is auto-committed with a descriptive message.
Best for: Terminal-first developers who want AI pair programming with automatic Git integration and multi-model flexibility.
CLI
Mode: CLI / Terminal·Pricing: Free
Proprietary
MCP4/5
Augment Code
Augment Code
The context depth champion, semantic Context Engine indexes 500K+ files across repositories, something no competitor matches at scale.
Best for: Enterprise teams with massive, multi-repository codebases who need cross-repo intelligence.
VS CodeJetBrainsCLI
Mode: Agentic·Pricing: Freemium
Open Source
MCP4/5
Cline
Community (originally Saoud Rizwan)
The most capable open-source VS Code coding agent with the broadest model support (10+ providers) and 5M+ installs.
Best for: Privacy-conscious developers who want agentic coding in VS Code with full control over model choice.
VS Code
Mode: Agentic·Pricing: Free
Proprietary
MCP4/5
Junie
JetBrains
Autonomous coding agent separate from JetBrains AI Assistant. Built-in test verification loop, LLM-agnostic CLI, and deep JetBrains IDE integration.
Best for: JetBrains IDE users and teams who want agentic coding with deep IDE integration and test-first verification.
JetBrainsCLI
Mode: Agentic·Pricing: Freemium
Proprietary
MCP4/5
Qodo
Qodo (formerly CodiumAI)
Quality-first AI tool with the highest F1 score (60.1%) in code review benchmarks. Complements rather than replaces general-purpose coding assistants.
Best for: Teams that need AI-powered code review, test generation, and quality assurance alongside their primary assistant.
VS CodeJetBrainsCLIWeb
Mode: Chat·Pricing: Freemium
Open Core
MCP4/5
Warp
Warp
The only tool that is simultaneously a production terminal emulator AND an AI agent orchestration platform. Block-based output, Oz cloud agents, and MCP support.
Best for: Terminal-heavy developers and DevOps engineers who want a modern terminal with built-in AI agents and cloud orchestration.
CLINative App
Mode: Terminal / ADE·Pricing: Freemium
Proprietary
MCP4/5
Windsurf
Cognition AI (formerly Codeium)
AI-native IDE with strong agentic capabilities via Cascade agent and persistent memory across sessions.
Best for: Developers who want a Cursor-like experience with strong agentic features and persistent memory.
Native AppVS Code
Mode: Agentic·Pricing: Freemium
Proprietary
MCP3/5
Amazon Q Developer
Amazon Web Services
Unmatched for AWS-native development with deep service integration, but general-purpose coding quality trails the leaders.
Best for: AWS-native teams who need deep integration with AWS services, IAM, CloudFormation, and the AWS SDK ecosystem.
VS CodeJetBrainsCLIWeb
Mode: Agentic·Pricing: Freemium
Proprietary
3/5
Blitzy
Blitzy, Inc. (Harvard Innovation Lab)
Not a copilot, a code factory. Orchestrates 3,000+ AI agents across 8-12 hours to batch-generate up to 3M lines of code from specifications.
Best for: Large enterprises with massive or legacy codebases (100M+ lines) who need autonomous bulk code generation.
VS CodeWeb
Mode: Code Factory·Pricing: Enterprise
Proprietary
3/5
Bolt.new
StackBlitz
Fastest time to working prototype (28 min average). Browser-based full-stack app generation for rapid prototyping.
Best for: Rapid prototyping and MVP generation for founders and designers.
Web
Mode: Browser-based·Pricing: Freemium
Open Source
MCP3/5
Continue
Continue Dev
Open-source IDE extension focused on model flexibility and customizability. Works in both VS Code and JetBrains.
Best for: Developers who want an open-source AI assistant in VS Code or JetBrains with full model control.
VS CodeJetBrains
Mode: Chat·Pricing: Free
Proprietary
MCP3/5
Google Antigravity
Google
Agent-first IDE from the $2.4B Windsurf acquisition. Manager View dispatches up to 5 parallel agents with built-in Chromium browser for testing.
Best for: Solo developers and frontend engineers who want multi-agent parallel execution with built-in browser testing.
Native App
Mode: Agentic·Pricing: Freemium
Proprietary
MCP3/5
Google Gemini Code Assist
Google
The most generous free tier (6,000 requests/day) and a massive 1M+ token context window, but response latency holds it back.
Best for: GCP-native teams and individual developers who want a capable free coding assistant with huge context capacity.
VS CodeJetBrainsCLIWeb
Mode: Chat·Pricing: Freemium
Proprietary
MCP3/5
JetBrains AI Assistant
JetBrains
The deepest IDE integration for JetBrains users. Inline coding helper (separate from Junie, the agentic agent).
Best for: Teams already committed to JetBrains IDEs who want native AI without switching editors.
JetBrains
Mode: Chat·Pricing: Freemium
Proprietary
3/5
Lovable
Lovable (formerly GPT-Engineer)
Design-first AI app builder with strong visual quality. 35 min average to working prototype.
Best for: Founders and designers who prioritise visual quality in AI-generated applications.
Web
Mode: Browser-based·Pricing: Freemium
Open Source
MCP3/5
OpenAI Codex CLI
OpenAI
OpenAI's terminal entry, built in Rust with 75K+ stars. Direct competitor to Claude Code in the CLI space.
Best for: Developers in the OpenAI ecosystem who want a terminal-first coding agent with GPT-5 integration.
CLI
Mode: CLI / Terminal·Pricing: Free
Open Source
MCP3/5
OpenCode
Community
Rising star of open-source terminal coding with 95K+ GitHub stars and 2.5M monthly developers. Built in Go for speed.
Best for: Terminal-first developers who want a fast, open-source alternative with a modern TUI.
CLI
Mode: CLI / Terminal·Pricing: Free
Open Source
MCP3/5
OpenHands
All-Hands-AI
The best-funded open-source autonomous coding agent ($18.8M Series A). MIT license, 71K+ GitHub stars, 53-77% SWE-bench.
Best for: Teams who want autonomous AI coding agents with full control, self-hosting, and no vendor lock-in.
WebCLI
Mode: Autonomous·Pricing: Free
Proprietary
MCP3/5
Replit Agent
Replit
The fastest path from idea to deployed app for non-developers. Revenue grew 10x after Agent launch.
Best for: Non-developers and founders who want to build and deploy working apps from natural language.
Web
Mode: Browser-based·Pricing: Freemium
Open Source
MCP3/5
Roo Code
Community
A Cline fork with additional features and its own community. Worth evaluating alongside Cline.
Best for: VS Code developers who want Cline-like capabilities with alternative development direction.
VS Code
Mode: Agentic·Pricing: Free
Proprietary
MCP3/5
Sourcegraph Amp
Sourcegraph (Amp — independent company)
Rebranded from Cody to Amp and spun out as independent company. Code search graph provides the deepest multi-repo understanding.
Best for: Large engineering organisations with complex multi-repository architectures.
VS CodeJetBrainsCLIWeb
Mode: Agentic·Pricing: Enterprise
Proprietary
MCP3/5
Tabnine
Tabnine
The privacy and compliance leader, SOC 2, HIPAA, GDPR, with on-premises and air-gapped deployment. Enterprise-only after killing the free tier.
Best for: Regulated industries (healthcare, finance, government) that need air-gapped AI coding with zero data retention.
VS CodeJetBrainsNeovim
Mode: Autocomplete·Pricing: Enterprise
Proprietary
3/5
v0
Vercel
The best AI tool for generating React/Next.js components with Tailwind CSS. Tightly coupled to the Vercel ecosystem.
Best for: Frontend developers building React/Next.js applications who want AI-generated components.
Web
Mode: Browser-based·Pricing: Freemium
// AI Coding Audit
NOT SURE WHICH
TOOL FITS YOUR WORKFLOW?
TOOL FITS YOUR WORKFLOW?
We map your development workflow, evaluate which AI coding tools fit your team size, stack, and security requirements, and give you a prioritised adoption plan.