Skip to content

AI coding agents

An AI coding agent reads and edits a codebase, runs commands and tests, and iterates on the result - going beyond code completion to carry out whole programming tasks.

Code completion suggests the next lines as you type. A coding agent takes a task ("fix this failing test", "add an endpoint"), explores the repository, edits several files, runs the tests and repeats until it works - usually with your approval for changes.

They run in different places: inside an editor, in the terminal, or as a cloud service that opens pull requests.

Examples tracked on AI Stats Live

AI coding agents: example tools with company, category, price and popularity rank
ToolCompanyCategoryFromFree tierPopularity
ChatGPTOpenAIGeneral assistants$8/moYes#3
ClaudeAnthropicGeneral assistants$20/moYes#4
Claude CodeAnthropicCoding agents$20/moNo#6
OpenAI CodexOpenAICoding agents$20/moNo#7
DeepSeekDeepSeekGeneral assistants—Yes#12
CursorAnysphereCoding$20/moYes#14
GitHub CopilotGitHubCoding$10/moYes#21
Replit AgentReplitApp builders$20/moYes#40
DevinCognitionCoding agents$20/moYes#47
opencodeSSTCoding agents—Unknown#67

Popularity is measured attention (Wikipedia, Hacker News, package downloads), not quality. Prices are the cheapest paid individual plan; each profile shows whether the price is verified. Data refreshed 1h ago.

Related ranking: Best for coding

  1. 1.GPT-6 Astra OpenAI74.1%
  2. 2.Gemini 3.8 Flash Google73.8%
  3. 3.Claude Opus 5 Anthropic73.6%
  4. 4.GPT-5.6 Sol OpenAI72.7%
  5. 5.Claude Fable 5 Anthropic69.9%

Ranked by DeepSWE - agentic software-engineering tasks on real repositories. Source: Epoch AI. Full ranking

Practical guides

Browse categories

Related terms

Explanation written by AI Stats Live editors; last reviewed 29 Sep 2026. Examples and rankings update automatically from the sources listed on the methodology page.