Viral post analysis
Score any X/social post on hook, structure and evidence — the exact judgment behind analysing 100k viral posts.
Movez ran the same narrow questions over 100,000 viral posts in 20.4 seconds for $0.67. The same corpus on a frontier chat model got through 214 posts and would have cost ~$458 — about 680× more per post. Viral analysis is the perfect Jev job: it's judging, not writing.
Try it live
This is the real thing, not a mockup. Edit the input, hit Run, and Jev returns every typed answer in one round trip — free, no signup. Now picture the same call fired across thousands of items in parallel.
Pick a demo, tweak the input, and hit Run Jev.
The decisions Jev makes
In a single call, Jev evaluates each of these — in parallel, against the same input:
What kind of hook does the first line use?
picks one of these options:
open_loop— creates curiosity/a gap the reader must resolvebold_claim— a big or contrarian claimstory— opens a personal narrativedata— leads with a number or statnone— no real hook
Does the hook open a curiosity loop the reader has to keep reading to close?
returns a calibrated yes/no probability.
Does the first line contain a specific number or dollar figure?
returns a calibrated yes/no probability.
How is the claim backed up?
picks one of these options:
demonstrated— shows real proof/receipts/specificsclaimed— asserts results but only claims themnone— no evidence at all
How strong is this post's viral potential?
rates it on an ordered scale:
- weak
- average
- strong
- very strong
The exact request
This is the real payload behind the live demo — copy it, change the state, and you're building:
{
"model": "jev-latest",
"state": "X post (12,400 likes, 1,900 reposts):\n\n\"I quit my $240k job with no backup plan.\n\nEveryone said I was insane.\n\n8 months later I'm making more than I ever did — and I only work 4 hours a day.\n\nHere's the exact 5-step system I used (steal it):\"",
"questions": {
"hook_type": {
"type": "choice",
"instructions": "What kind of hook does the first line use?",
"criteria": {
"open_loop": "creates curiosity/a gap the reader must resolve",
"bold_claim": "a big or contrarian claim",
"story": "opens a personal narrative",
"data": "leads with a number or stat",
"none": "no real hook"
}
},
"opens_loop": {
"type": "noul",
"instructions": "Does the hook open a curiosity loop the reader has to keep reading to close?"
},
"first_line_number": {
"type": "noul",
"instructions": "Does the first line contain a specific number or dollar figure?"
},
"evidence": {
"type": "choice",
"instructions": "How is the claim backed up?",
"criteria": {
"demonstrated": "shows real proof/receipts/specifics",
"claimed": "asserts results but only claims them",
"none": "no evidence at all"
}
},
"virality": {
"type": "score",
"instructions": "How strong is this post's viral potential?",
"criteria": [
"weak",
"average",
"strong",
"very strong"
]
}
}
}Wire it into your code
Read the typed answers and branch in plain code — no parsing. Auto-handle the high-confidence cases and route the uncertain ones to a bigger model or a human. It's one API call and output is free, so ask every question you need at once.
Build your own
Every scenario above is a single API call. Try any of them free in the playground, then get a hosted key to ship it in minutes.