Launching today
ParaPulse
See which AI models are rising — before the crowd catches on
1 follower
See which AI models are rising — before the crowd catches on
1 follower
ParaPulse tracks momentum across thousands of open-source AI models and datasets — not just raw download counts, but growth rates, moving averages, and trend rankings. See who's rising, who's fading, and how fast: - Week-over-week & month-over-month growth rates - Momentum-weighted leaderboards - Trend charts (7d / 30d / 90d) - Multi-model comparison - Public REST API Free. No login. No paywall.
Hey Product Hunt! I'm the builder behind ParaPulse, and I want to share the itch it scratches.
Every time I evaluated a new AI model, I'd hit the same wall: the only number available was total downloads — a cumulative stat that tells you about the past, not the present. A model could be coasting on a viral launch from 6 months ago, or quietly doubling its adoption week over week, and you'd never know.
So I built ParaPulse to answer the questions that actually matter:
- Is this model gaining or losing momentum right now?
- How does it compare to its direct competitors over the last 30 days?
- Which categories (embeddings, LoRA adapters, vision models…) are heating up vs. cooling off?
The stack: daily snapshots → 7-day moving average → WoW/MoM growth rates → momentum-weighted scores → precomputed leaderboards served fast.
Everything is public and read-only — no account, no API key, no paywall. There's also a REST API if you want to pull trend data into your own tooling.
I'd love feedback from this community on two things:
1. Which metrics matter most when you're deciding whether to adopt a new model?
2. Are there categories or use cases (fine-tuning datasets, multimodal models, reasoning models) you'd want leaderboards for specifically?