Agent · Autonomous
Devin
Devin is Cognition's autonomous software engineer that plans, writes, runs, and debugs code end-to-end in its own sandbox with a full browser, editor, and terminal.
by Cognition·Closed product·Launched Mar 2024
Capability
—
no benchmarks
GitHub stars
—
Harness
Plan + execute
Built-in tools
Environments
2
Cloud · Web
Capability
No public benchmark submissions · yet
Devin has not published benchmark scores on SWE-bench Verified, GAIA, OSWorld or τ-bench. Check the GitHub repo or vendor site for the latest · we track new submissions daily.
The Stack
Brain · Harness · Tools · Environment
Brains
Which LLM powers it
claude-opus-4-6
Harness
The loop pattern
Plan + execute
Tools
How it acts on the world
Built-in
Environments
Where it lives
CloudWeb
Frequently asked
Pulled from the dataset · updated daily
What is Devin?
Devin is Cognition's autonomous software engineer that plans, writes, runs, and debugs code end-to-end in its own sandbox with a full browser, editor, and terminal.
What model powers Devin?
Devin runs on claude-opus-4-6. It uses a plan + execute pattern with built-in tools.
Is Devin open-source?
No · Devin is a closed product from Cognition. You can use it via their subscription.
Also in Autonomous
Peer agents in the same category · sorted by capability