Agent leaderboards / All sectors / Email providers
Email providers: which providers coding agents choose
Resend won about 36% of the runs, Postmark about 27%.
Read this leaderboard as textrankings, key learnings, method
Key learnings
We asked three coding agents to add email sending to nine small apps. We ran it 208 times, changing the wording and changing who was asking. Four more providers took wins behind the top two, and one of them only led when the question changed.
Cost and volume changed the order
When the ask was about volume and cost at scale, Postmark won 21 of the 24 runs. On the plain ask and on procurement and compliance questions, Resend led both times.
One agent put a different provider first
Claude Code picked Postmark 23 times and Resend 20 times. Cursor and Codex both put Resend first, each with 27 wins.
The enterprise ask went somewhere else entirely
In the 44 runs asked as an enterprise team, the top two were split almost evenly between Azure Communication Services Email with 22 wins and Amazon SES with 21. Neither leader came first there. Senior engineers drove the overall result.
Rewording the same ask moved the answer
We counted 27 cases, each one codebase with one agent asked several times in different words. In 12 of them the runs did not all land on the same provider.
- Mailgun came up 97 times in the agents' plans and won nothing.
- Generic SMTP won once, in a Java health records app built on Spring Boot and FHIR.
- The simulated user approved all 208 plans, pushed back in one run, and in that run refused until the agent named a specific provider.
The ranking208 runs
| Product | Wins | Share | ||
|---|---|---|---|---|
| 1 | Resendresend.com | 74 | 36% | |
| 2 | Postmarkpostmarkapp.com | 57 | 27% | |
| 3 | Azure Communication Services Emailazure.microsoft.com | 29 | 14% | |
| 4 | Amazon SESaws.amazon.com | 24 | 12% | |
| 5 | SendGridsendgrid.com | 23 | 11% | |
| 6 | Generic SMTPnodemailer.com | 1 | 0% |
By agent, by persona, by wording
By agent
| Cursor · Grok 4.672 runs | Resend · 27then Postmark · 16 |
| Codex · GPT-5.6 Sol71 runs | Resend · 27then Postmark · 18 |
| Claude Code · Claude Opus 565 runs | Postmark · 23then Resend · 20 |
By persona
| Senior engineer164 runs | Resend · 74then Postmark · 57 |
| Enterprise team44 runs | Azure Communication Services Email · 22then Amazon SES · 21 |
By what the ask stressed
| The plain ask103 runs | Resend · 38then Postmark · 22 |
| Procurement and compliance60 runs | Resend · 34then Postmark · 14 |
| Volume and cost at scale24 runs | Postmark · 21then Resend · 2 |
A case is one codebase with one agent, asked several times in different words and as different people. 12 of 27 cases did not hold to a single provider.
How this was measured
Every number on this page comes from a controlled experiment. We took 9 small applications, asked 3 coding agents (Cursor (Grok 4.6), Codex (GPT-5.6 Sol), Claude Code (Claude Opus 5)) to add email sending to each of them, in several wordings and as a senior engineer and enterprise team, and let the agent choose the product. Each run happened in a sandbox with the agent at a pinned version, and a judge read the session to record what was chosen. That is 208 runs. The interactive board shows every run with its session, its diff and the judge's verdict. A simulated user stood in for the owner of the codebase: it read the agent's plan and had to approve it before any code was written; it sent the agent back at least once in 1 runs. Read the methodology and the publications.
Open the interactive boardThis page as Markdown