think-reason-learn/linkedin-viral-post-predictorPredicts whether a LinkedIn post will be a hit or go viral, and explains what pulled the score up or down.
TextImage$0.01 / request2.00 s
NEWScore posts against the author's own account
Prediction models that show their reasons
Pretrained judges for questions like will this post do well? and does this design look generated? Each call returns a calibrated score and what the model noticed, so a coding agent can generate, score and improve in a loop.
OpenAI-compatible. Also available on OpenRouter.
import json
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.thinkreasonlearn.com/v1",
api_key=os.environ["TRL_API_KEY"],
)
response = client.chat.completions.create(
model="think-reason-learn/linkedin-viral-post-predictor",
messages=[
{
"role": "user",
"content": "We almost ran out of money in month 9.\n\nHere's the one decision that saved the company — and why I'd make it again:\n\nWe stopped selling to everyone and picked 1…",
}
],
)
prediction = json.loads(response.choices[0].message.content)
print(prediction)Example response
A score your agent can act on
The LinkedIn Viral Post Predictor read a draft and found three things working and one holding it back. Move the link to the comments, score again, and keep the better version.
Cheap enough to call on every draft
About $0.01 per prediction. Score twenty variants of a post for 20 cents.
Fast enough for a loop
A trained scorer combines hundreds of plain-language checks. A prediction takes a few seconds.
Shows its reasons
Every score comes with the checks that pushed it up or down, so your agent knows what to fix.
90-98th
Scores in the 90-98th percentile of posts by authors with a similar following. Three checks pulled it up and one pulled it down.
84%
Content score
0%100%
Probabilities
Hit (top 5%)14%2.8× base
Viral (top 1%)3%3.1× base
Bars run from 0 to 18%. Ticks mark the base rate.
What the model noticed
Pushed up3
- Does the first line work as a hook on its own?.72
- Is it told as a personal story?.55
- Does it set up an expectation and then turn it around?.48
Pushed down1
- Does it send readers off LinkedIn?.47
Models
Trained with Think-Reason-Learn on real outcomes.
think-reason-learn/linkedin-viral-post-predictor-with-accountScores a LinkedIn post against the author's own account and audience, and shows what the account changes.
TextImagefrom $0.01 / request3.00 s
think-reason-learn/github-pr-merge-predictorPredicts whether a pull request from an outside contributor will be merged, and explains what pulled the score up or down.
Text$0.01 / request3.00 s
think-reason-learn/github-contributor-ratingRates how likely a GitHub user's pull requests to other people's repos are to be merged, from their public track record.
Text$0.01 / request5.00 s
think-reason-learn/show-hn-launch-predictorPredicts whether a Show HN post will reach 100+ points on Hacker News, and explains what pulled the score up or down.
Text$0.01 / request700 ms
AI Slop Design Detector
Coming soonthink-reason-learn/ai-slop-design-detectorScores a web page design for craft and tells you what makes it look AI-generated, with the fixes that matter most.
TextImage$0.10 / request
X Quote & Reply Predictor
Coming soonthink-reason-learn/x-quote-reply-predictorPredicts how much conversation a post on X will start: quotes, replies and reach.
TextImage$0.01 / request
Pay as you go, from $5
Add credits when you need them. No subscription, no seats.