GitHub Spark Experimental AI App Builder to Be Deprecated in August 2026

GitHub has officially announced the sunset of GitHub Spark, its experimental tool designed for building web applications using natural language. Effective August 4, 2026, new user registrations and new application creation have been disabled. Active developers can continue to access their existing projects to export code and configurations until the complete platform shutdown on August 31, 2026. The deprecation aligns with GitHub's focus on integrating agentic capabilities directly into existing development tools like GitHub Copilot and Visual Studio Code. Underpinning this change, the GitHub Models inference service, which powered the native llm() function in Spark, was terminated on July 30, 2026. Consequently, any applications relying on this specific function for AI capabilities have already ceased functioning. Applications built on Spark that do not utilize the native llm() function will remain active and editable until the final shutdown date. Developers who wish to preserve their application logic must export their source code and migrate external dependencies, such as switching to alternative LLM API providers if their projects rely on runtime AI model inference. Deployed client-side builds of Spark applications will continue to run externally after the platform is retired.
Related tools
Recommended tools for this topic
These picks prioritize high-intent tools relevant to this topic. Some links may include partner or affiliate tracking.
Strong fit for AI, backend, and frontend readers looking for an AI-first coding workflow.
View CursorHigh-value hosting and deployment path for frontend and cloud readers.
View VercelA strong security and edge platform match across CDN, Zero Trust, and app protection.
View CloudflareComparison
| Aspect | Before / Alternative | After / This |
|---|---|---|
| User Onboarding | Open registration and creation of new Spark applications | Registrations and new project creation fully disabled |
| GitHub Models API | Native llm() function enabled via GitHub Models | Terminated on July 30, 2026; requires third-party API keys |
| Workspace Access | Full access to edit, run, and export code in the dashboard | Complete service shutdown on August 31, 2026 |
Action Checklist
- Export all application code from the GitHub Spark dashboard before August 31, 2026. Ensure all files are saved locally as no cloud backups will be accessible after this date.
- Identify and replace any native llm() function calls in your application logic. These functions have been non-operational since the GitHub Models shutdown on July 30, 2026.
- Integrate alternative LLM provider APIs if runtime inference is required. You will need to configure custom API keys for services like OpenAI or Anthropic.
- Verify external hosting arrangements for your deployed client-side applications. Deployed frontend builds will continue to function independently of the Spark platform.
Source: GitHub Changelog
This page summarizes the original source. Check the source for full details.

