Loading the page.
AI IDE and editor ratings
Development environments and editors where AI is part of the daily workflow: Cursor, Antigravity, and similar tools.
Leadership: Cursor 90 · Opus —·Calculation: source consensus
Market leader
Cursor
90/100Average score
51
/100Average coverage
50%
sourcesHigh confidence
1
6 assetsCoverage
5
sources| # | Name | Valuation | Coverage | Confidence | Value | Best for | Sources | Profile |
|---|---|---|---|---|---|---|---|---|
| 1 |  CursorCursorOpen Cursor | 90 | 50% 2/4 | Medium 66/100 | reliable daily tool | daily IDE development: edits, navigation, agent mode, worktrees, teams where iteration speed matters more than full autonomy | ||
| 2 |  WindsurfWindsurfOpen Cognition | 77 | 75% 3/4 | High 81/100 | strong IDE layer | IDE-first development with agentic scenarios, teams where Cursor does not fit due to UX or policy | ||
| 3 |  Google AntigravityAntigravityOpen | 64 | 50% 2/4 | Medium 66/100 | niche UI tool | experiments with agent-first IDE and browser primitives, Google/Gemini-oriented workflow | ||
| 4 |  JetBrains JunieJunieOpen JetBrains | 32 | 50% 2/4 | Medium 68/100 | IDE-native | IntelliJ/PyCharm/WebStorm teams, enterprise IDE workflow and JetBrains AI ecosystem | ||
| 5 |  Augment CodeAugmentOpen Augment | 28 | 50% 2/4 | Medium 68/100 | large-system context | large codebase context, enterprise controls, IDE/CLI workflow, teams where understanding the whole repository matters | ||
| 6 |  ClineClineOpen Cline | 13 | 25% 1/4 | Medium 45/100 | community value | open-source IDE agent workflow, BYOK, Cline credits, approval-first changes |
Methodology V0
The LLM ranking is calculated only from public coding signals: DeepSWE is required as an anchor and receives 60%, SWE-Rebench 20%, Artificial Analysis 10%, Aider 10%. Price and editorial caveats are shown separately. For AI agents, a separate V0 view is currently shown based on available sources and confidence.
How the index is calculated