Agentic AI Landscape and Trends

🧐 Latest Blog: Agentic AI 2026: When the Hackathon Fever Cools Down

Agentic AI Landscape 2026

The 2026 landscape maps two infrastructure blocks: Agent Infra organizes the application, framework, runtime, and tool ecosystem; Model Infra covers the data, training, serving, and deployment stack.

The landscape highlights the projects that are currently most representative of each ecosystem rather than attempting to cover every project. Visit InclusionAI Insights for more complete and dynamic Agentic AI ecosystem rankings, project data, and developer details.

Insights

• Landscape reports — dated, bilingual analyses of how each layer of the ecosystem is evolving.
• Case studies — deep dives into single projects or themes.
• Weekly reports — automatically generated snapshots of newly surfacing projects.

Data

The canonical dataset is data/agentic-ai-projects.csv. Each row is keyed by the GitHub repo_id, carries GitHub metadata (stars, forks, license, language, topics), OpenDigger signals (openrank_*, participants_*), and the curation fields that record why a project is in the landscape (landscape_layer, landscape_section, selection_reason, selection_caveat).

Project vitality is measured with OpenRank rather than raw star counts, so activity from issues, pull requests, reviews, and contributors is taken into account.

Landscape Website

The production Next.js application lives in apps/landscape-web and reads the canonical project dataset above directly.

For a local preview with live updates, run this from the repository root:

npm run web:dev

Open http://127.0.0.1:3000. Changes under apps/landscape-web/ and the canonical CSV will be rebuilt automatically. This command uses Webpack polling because native file watching has previously exhausted file handles in this repository.

If the page was already open against npm run start, refresh it once after switching servers so the browser loads the development client. After that, changes appear without another manual refresh.

Use npm run start only when checking a completed production build; it does not provide live updates.

The existing production address is landscape-demo-omega.vercel.app. Vercel should use apps/landscape-web as the project Root Directory.

Maintaining The Agentic AI Projects

We aim to continuously maintain and expand data/agentic-ai-projects.csv with noteworthy projects across the Agentic AI ecosystem. If you think an important project is missing, please share it through our dedicated issue tracker.

The data collection and publishing code lives in scripts/. To run it locally:

python3 -m venv .venv
.venv/bin/pip install -r requirements.txt
# then fill in scripts/.env with GitHub, ClickHouse, and publishing credentials

Weekly report and ecosystem insight operations are documented in WORKFLOW.md; repository conventions for contributors and coding agents are in AGENTS.md.

Initiated by Communities