Replit Agent
Replit Agent scaffolds, edits and deploys full applications from natural language, billed by effort-based agent checkpoints.
Full Replit Agent record →Head to head
Replit Agent vs Claude Code
Replit Agent (Replit) and Claude Code (Anthropic) both sit in Coding. Claude Code shipped more tracked changes in the last 30 days (10 vs 6). Every value below comes from the latest crawl of the vendors' own pages.
| Attribute | Replit Agent Replit | Claude Code Anthropic |
|---|---|---|
| Segment | Coding | Coding |
| Version | Replit Agent | latest |
| Entry cost | Free tier · from $20/mo | from $20/mo |
| Pricing tiers | Core free · Pro free | Pro $20 · Team (Standard seat) $25 · Team (Premium seat) $125 · Enterprise $20 |
| Model stack | Agent Max Models · Agent Power Models · Claude · Free Mode · Free Mode Model · frontier models · GPT-5 class · Max Mode · Max Mode Model · Max mode models · Power Mode · Power Mode Model · Power mode models · unspecified frontier model · unspecified frontier models | Claude 3.5 Sonnet · Fable 5 · Fable 5.1 · Haiku · Haiku 4.5 · Haiku 5.5 · Opus · Opus 5 · Opus 5.5 · Sonnet · Sonnet 3.5 · Sonnet 5 · Sonnet 5.5 |
| Context window | Model-dependent | Up to 1M varies by model |
| Public API | ||
| Routes models | ||
| Changes / 30d | 6 | 10 |
| Origin | Global | Global |
| Capabilities | Prompt to deployed app, Autonomous testing loop, Database and auth provisioning, Mobile app, Checkpoint billing, Plan mode, Lite mode, Economy mode | Terminal agent, Repo-wide edits, Subagents, MCP servers, Hooks, Background tasks, Agentic coding CLI, VS Code extension |
Claude Code
A terminal-native coding agent that reads, edits and runs code across a whole repository.
Full Claude Code record →