Skip to content

Resend vs SendGrid: what coding agents actually pick

Resend got installed in 11 of 36 email runs across Opus 5.5, Sonnet 5.5 and Haiku 5.5 (31%). SendGrid got installed in 0. the most common result overall was neither: Raw SMTP, 24 of 36.

SendGrid is the long standing email API. Resend is the newer one. we asked Claude Code to add email to a blank app without naming a provider and counted what it installed.

36 scored email runsOpus 5.5, Sonnet 5.5 and Haiku 5.5 on Claude Codeupdated Oct 18, 2026all email results

Resend vs SendGrid, side by side

runs where the agent installed or wired in the tool. named means it showed up in the final message but not in the code.

toolinstalledOpus 5.5Sonnet 5.5Haiku 5.5Next.js + TypeScriptPython + FastAPInamed only
Resend11 of 36 (31%)4 of 122 of 125 of 1211 of 180 of 184
SendGrid0 of 36 (0%)0 of 120 of 120 of 120 of 180 of 189
Postmark1 of 36 (3%)1 of 120 of 120 of 120 of 181 of 1811
Raw SMTP24 of 36 (67%)7 of 1210 of 127 of 127 of 1817 of 180

how this was measured

this is not a feature review. it counts what Claude Code reaches for when the prompt leaves the choice to it. every run starts from the same blank starter repo, Next.js + TypeScript or Python + FastAPI. the prompt asks for the feature and never names a tool. a pick is whatever ends up installed or wired into the code, read from the diff.

the 3 prompts we rotate through:

  1. “The app needs to send a welcome email when someone signs up. Wire up email sending and add the template.”
  2. “Add transactional email to this project: a password reset message and a receipt message, sent from the backend.”
  3. “I need to email users from this app. Set up whatever is needed to send mail in production and give me one function I can call.”

get told if SendGrid catches up

new models get the same email prompts. one email with what moved. free. read issue 1

every email result

all models pooled, 36 scored runs. most common: Raw SMTP, 24 of 36.

  1. Raw SMTP24 of 3667%
  2. Resend11 of 3631%
  3. Postmark1 of 363%
  4. built its own, any form24 of 3667%
Which email API do coding agents pick? chart from What Agents Pick

other head to heads

how a pick gets counted: every run starts from the same blank starter repo, the prompt never names a tool, and a pick is whatever the agent installs or wires in, read from the diff. full rules on the methodology page, every run for email on the category page, raw data in runs.json.