2026年9月
Replit introduced Replit Agent, a new agentic architecture that lets frontier models like GPT-6 Astra dynamically delegate tasks to subagents based on cost and
Replit launched a Model selector for its Agent feature, enabling users to manually choose models or rely on intelligent routing that balances cost and quality.
Replit introduced 14 new Model Context Protocol (MCP) server integrations for its Replit Agent, enabling one-click installation of tools for Airtable, Airwallex
Replit announced the acquisition of Atta to integrate its business analysis and high-quality charting capabilities directly into Replit chat. The first feature,
Replit introduced custom API connectors in beta for Pro and Enterprise customers, allowing workspaces to integrate unsupported APIs directly. Admins can configu
Replit introduced beta support for custom API connectors in Agent for Pro and Enterprise teams, enabling integration with external APIs beyond its library. Ente
Replit introduced an MCP server enabling AI tools like ChatGPT, Claude, and Slack to create, inspect, update, and publish Replit apps directly. The integration
Replit introduced three new features: desktop MCP client integration for app creation and management, managed PostgreSQL via Lakebase in Databricks Apps for sea
Replit announced the general availability of its integration with Databricks, now featuring native Lakebase support. This allows enterprise teams to build produ

Replit opened its first international office in London to serve its growing European user base, including 85% of the Fortune 500. The company plans to double it

Replit introduced Project Analytics in the Growth pane to track visitor trends, traffic sources, and custom events for published web apps. Paid plans unlock cus
Replit introduced point-in-time restore and scheduled backups for production databases, extending recovery windows from 7 to 28 days for Pro and Enterprise plan
Replit introduced scheduled backups for production databases, allowing daily restore points with configurable retention periods based on plan. It also launched
2026年8月
Replit introduced intelligent model routing in its chat feature, allowing the platform to automatically select the most suitable AI model for each task based on
Replit introduced intelligent model routing in Agent, automatically selecting models and effort levels to balance quality and cost. Enterprise Workspaces now de
Replit introduced Intelligent Model Routing, an AI feature that automatically selects the best model for each task to balance quality, speed, and cost. In testi

归档中还有 +227 条更新:可按上方的月份或年份浏览。
自动跟踪 Replit