AI Deploy Bench

Trends: Static site

Each metric over runs, one line per cloud. Hover over a point for its value, and select a cloud in a legend to hide or show it.

Every run that included the static site, averaged over trials.

Success rate

Share of apps whose deployment passed every required check. Higher is better.

0%25%50%75%100%Oct 13 #4Oct 13 #5Oct 13 #8Oct 14 #9Cloudish · Oct 13 #4: 100%Cloudish · Oct 13 #5: 100%Cloudish · Oct 13 #8: 100%Cloudish · Oct 14 #9: 100%DigitalOcean · Oct 13 #4: 100%DigitalOcean · Oct 13 #5: 100%DigitalOcean · Oct 13 #8: 100%DigitalOcean · Oct 14 #9: 100%Render · Oct 13 #4: 100%Render · Oct 13 #5: 100%Render · Oct 13 #8: 100%Render · Oct 14 #9: 100%Vercel · Oct 13 #4: 100%Vercel · Oct 13 #5: 100%Vercel · Oct 13 #8: 100%Vercel · Oct 14 #9: 100%

Checks passed

Share of the verifier's checks that passed, averaged over apps. Higher is better.

0%25%50%75%100%Oct 13 #4Oct 13 #5Oct 13 #8Oct 14 #9Cloudish · Oct 13 #4: 100%Cloudish · Oct 13 #5: 100%Cloudish · Oct 13 #8: 100%Cloudish · Oct 14 #9: 100%DigitalOcean · Oct 13 #4: 100%DigitalOcean · Oct 13 #5: 100%DigitalOcean · Oct 13 #8: 100%DigitalOcean · Oct 14 #9: 100%Render · Oct 13 #4: 100%Render · Oct 13 #5: 100%Render · Oct 13 #8: 100%Render · Oct 14 #9: 100%Vercel · Oct 13 #4: 100%Vercel · Oct 13 #5: 100%Vercel · Oct 13 #8: 100%Vercel · Oct 14 #9: 100%

Human steps

Things a person had to do, or would have to do, for the agent to deploy. Lower is better.

01234Oct 13 #4Oct 13 #5Oct 13 #8Oct 14 #9Cloudish · Oct 13 #4: 0Cloudish · Oct 13 #5: 0Cloudish · Oct 13 #8: 0Cloudish · Oct 14 #9: 0DigitalOcean · Oct 13 #4: 3DigitalOcean · Oct 13 #5: 3DigitalOcean · Oct 13 #8: 3DigitalOcean · Oct 14 #9: 3Render · Oct 13 #4: 2Render · Oct 13 #5: 2Render · Oct 13 #8: 2Render · Oct 14 #9: 2Vercel · Oct 13 #4: 2Vercel · Oct 13 #5: 2Vercel · Oct 13 #8: 2Vercel · Oct 14 #9: 2

Human time

Estimated minutes those human steps take. Lower is better.

0m2.5m5m7.5m10mOct 13 #4Oct 13 #5Oct 13 #8Oct 14 #9Cloudish · Oct 13 #4: 0 minCloudish · Oct 13 #5: 0 minCloudish · Oct 13 #8: 0 minCloudish · Oct 14 #9: 0 minDigitalOcean · Oct 13 #4: 7 minDigitalOcean · Oct 13 #5: 7 minDigitalOcean · Oct 13 #8: 7 minDigitalOcean · Oct 14 #9: 7 minRender · Oct 13 #4: 5 minRender · Oct 13 #5: 5 minRender · Oct 13 #8: 5 minRender · Oct 14 #9: 5 minVercel · Oct 13 #4: 5 minVercel · Oct 13 #5: 5 minVercel · Oct 13 #8: 5 minVercel · Oct 14 #9: 5 min

Security & compliance

Score of the independent security and compliance review (0-100). Higher is better.

0255075100Oct 13 #4Oct 13 #5Oct 13 #8Oct 14 #9Cloudish · Oct 13 #5: 83/100Cloudish · Oct 13 #8: 75/100Cloudish · Oct 14 #9: 82/100DigitalOcean · Oct 13 #5: 80/100DigitalOcean · Oct 13 #8: 89/100DigitalOcean · Oct 14 #9: 91/100Render · Oct 13 #5: 74/100Render · Oct 13 #8: 72/100Render · Oct 14 #9: 72/100Vercel · Oct 13 #5: 73/100Vercel · Oct 13 #8: 74/100Vercel · Oct 14 #9: 76/100

Build time

Wall-clock time of the agent's build phase, averaged over apps. Lower is better.

0s50s100s2m3mOct 13 #4Oct 13 #5Oct 13 #8Oct 14 #9Cloudish · Oct 13 #4: 2m 06sCloudish · Oct 13 #5: 1m 18sCloudish · Oct 13 #8: 1m 17sCloudish · Oct 14 #9: 1m 47sDigitalOcean · Oct 13 #4: 1m 31sDigitalOcean · Oct 13 #5: 1m 26sDigitalOcean · Oct 13 #8: 1m 28sDigitalOcean · Oct 14 #9: 1m 45sRender · Oct 13 #4: 1m 21sRender · Oct 13 #5: 1m 20sRender · Oct 13 #8: 1m 45sRender · Oct 14 #9: 1m 20sVercel · Oct 13 #4: 1m 25sVercel · Oct 13 #5: 1m 14sVercel · Oct 13 #8: 1m 22sVercel · Oct 14 #9: 1m 16s

Deploy time

Wall-clock time of the agent's deploy phase, averaged over apps. Lower is better.

0s2m4m6m8mOct 13 #4Oct 13 #5Oct 13 #8Oct 14 #9Cloudish · Oct 13 #4: 6m 53sCloudish · Oct 13 #5: 3m 38sCloudish · Oct 13 #8: 2m 41sCloudish · Oct 14 #9: 2m 12sDigitalOcean · Oct 13 #4: 2m 26sDigitalOcean · Oct 13 #5: 2m 18sDigitalOcean · Oct 13 #8: 1m 38sDigitalOcean · Oct 14 #9: 1m 54sRender · Oct 13 #4: 2m 08sRender · Oct 13 #5: 2m 44sRender · Oct 13 #8: 1m 52sRender · Oct 14 #9: 1m 22sVercel · Oct 13 #4: 1m 57sVercel · Oct 13 #5: 1m 49sVercel · Oct 13 #8: 1m 04sVercel · Oct 14 #9: 1m 02s

Tokens

Tokens the agent used, including cache reads and writes; totals across apps. Lower is better.

0500.0K1.00M1.50M2.00MOct 13 #4Oct 13 #5Oct 13 #8Oct 14 #9Cloudish · Oct 13 #4: 1.02MCloudish · Oct 13 #5: 726.0KCloudish · Oct 13 #8: 833.2KCloudish · Oct 14 #9: 1.21MDigitalOcean · Oct 13 #4: 484.7KDigitalOcean · Oct 13 #5: 646.6KDigitalOcean · Oct 13 #8: 650.9KDigitalOcean · Oct 14 #9: 732.6KRender · Oct 13 #4: 537.7KRender · Oct 13 #5: 1.16MRender · Oct 13 #8: 866.1KRender · Oct 14 #9: 643.6KVercel · Oct 13 #4: 419.6KVercel · Oct 13 #5: 594.9KVercel · Oct 13 #8: 421.7KVercel · Oct 14 #9: 547.9K

Agent cost

Model API cost of the build and deploy sessions; totals across apps. Lower is better.

$0$0.25$0.5$0.75$1Oct 13 #4Oct 13 #5Oct 13 #8Oct 14 #9Cloudish · Oct 13 #4: $0.73Cloudish · Oct 13 #5: $0.52Cloudish · Oct 13 #8: $0.57Cloudish · Oct 14 #9: $0.73DigitalOcean · Oct 13 #4: $0.45DigitalOcean · Oct 13 #5: $0.48DigitalOcean · Oct 13 #8: $0.61DigitalOcean · Oct 14 #9: $0.51Render · Oct 13 #4: $0.50Render · Oct 13 #5: $0.81Render · Oct 13 #8: $0.76Render · Oct 14 #9: $0.61Vercel · Oct 13 #4: $0.42Vercel · Oct 13 #5: $0.57Vercel · Oct 13 #8: $0.49Vercel · Oct 14 #9: $0.54

Hosting, light use

Estimated monthly hosting cost of the deployed resources: barely used: a few visits a week, about 2,000 requests a month. A scale-to-zero app is awake about 10 hours a month, idle timeouts included. Lower is better.

$0$6.25$12$19$25Oct 13 #4Oct 13 #5Oct 13 #8Oct 14 #9Cloudish · Oct 13 #8: $0.11Cloudish · Oct 14 #9: $0.11DigitalOcean · Oct 13 #8: $5.00DigitalOcean · Oct 14 #9: $5.00Render · Oct 13 #8: $0.00Render · Oct 14 #9: $0.00Vercel · Oct 13 #8: $20.00Vercel · Oct 14 #9: $20.00

Hosting, busy

Estimated monthly hosting cost of the deployed resources: a public app that crawlers, uptime monitors and visitors hit around the clock: one request a second on average, so it never goes idle. Lower is better.

$0$6.25$12$19$25Oct 13 #4Oct 13 #5Oct 13 #8Oct 14 #9Cloudish · Oct 13 #8: $7.84Cloudish · Oct 14 #9: $7.84DigitalOcean · Oct 13 #8: $5.00DigitalOcean · Oct 14 #9: $5.00Render · Oct 13 #8: $7.13Render · Oct 14 #9: $7.13Vercel · Oct 13 #8: $20.00Vercel · Oct 14 #9: $20.00