Coding assistants
Devin
Made by Cognition
Cognition's autonomous coding agent — listed for its free documentation and free tier.
What it is
Devin aims higher than the other coding tools here: give it a whole ticket and it tries to finish the ticket, including running the code and fixing what breaks. On a well-scoped task that is impressive. When it misunderstands the task, it spends your money exploring.
I list it mostly for the documentation, which is free and is one of the clearest public explanations of how an autonomous coding agent is actually designed.
There is a free tier with a light quota. The useful quotas are paid, so treat this as reading material rather than a free tool.
Listed here for the free documentation and free tier — the useful quotas are paid.
Published price
- FreeLight quota, limited models$0/month
- Pro$20/month
- Max$200/month
- Teams$80/month + $40 per developer seat
- EnterpriseCustom, billed in compute units
Source: Devin's pricing page · Last checked 17 September 2026
Benchmarks
The published figures for Devin, copied exactly as their sources give them and checked 17 September 2026. Each one names the test, who ran it and when, and says when the number comes from the maker rather than an independent tester.
- SWE-bench (original, unassisted)13.86%
share of real GitHub issues resolved end to end
Cognition's own technical report · 15 March 2024 · maker's own figures
The maker's own figure, and now two years old.
- SWE-bench Verified48.2% or 61.7%, depending on the source
share of human-checked GitHub issues resolved end to end
Dataku (48.2%) and Tensorfeed (61.7%) · 9 June 2025 / undated · independent
The two aggregators contradict each other and neither is the official leaderboard. Treat as unverified.
Devin's numbers are a mess to cite honestly: Cognition's only first-party report predates the SWE-bench Verified set, and the two aggregators that list a current figure disagree by 13 points. Both are below.
Figures copied from the sources above, checked 17 September 2026.
Where to read today's numbers
Scores move, so these are the boards that keep them current.
- SWE-bench leaderboardIndependent
Measures whether an agent can fix real GitHub issues. The standard reference for coding agents.
- Cognition's own benchmark write-upsMaker's own figures
The maker's own results, including its original agent-benchmark claims. Cross-check against SWE-bench.
How it performs, in my experience
Aims to take a whole ticket end to end rather than answer one question. Impressive on well-scoped tasks, expensive when it wanders.
Free value
1 / 5
How much you get without paying.
Performance
4 / 5
How well it does its main job.
Range of uses
2 / 5
How many different jobs it suits.
Listed for its free documentation — the product itself is a paid subscription. These scores are my own judgment, not a measurement — the benchmark links above are the independent version.
Prompting it from this site
Can't be run from here
Runs only inside Cognition's own cloud, on whole projects.
Use it for
- Understanding how autonomous coding agents are designed
Running it yourself
Hosted only — nothing to install
Devin is a closed, hosted product. It runs in the maker's own cloud environment, and there is nothing to download or install.
What you need first
- A paid Cognition account
Step by step
1.Use it in the browser or through its Slack app — nothing to install
If you want a coding agent that runs on your own machine against your own files, OpenCode or Kilo are the local equivalents.