Forget README claims. Models exported from Model Surgeon ship with a cryptographically signed surgical history sealed to the file's fingerprints.
Anyone can verify your model's integrity offline for free with our open-source verifier. Change one byte, and it flips to TAMPERED.
Cut a layer out of GPT-2's brain. Live. In your browser.
Model Surgeon opens most AI models as 3-D anatomy — excise, graft, merge, quantize — and every export ships with a signed lineage anyone can verify.
Nothing to install. Looking is free.
https://t.co/MvooHpT3IX
You'd never deploy code you can't inspect. Why deploy models you can't?
Model Surgeon: open any checkpoint in-browser, scan for tampering, prune with evidence, export with a signed record anyone can verify — free.
Visual. No-code. Zero-execution safe.
https://t.co/khMmgELGUS
1.27 billion total downloads for Falconsai/nsfw_image_detection. We’re sitting at #3 all-time on HuggingFace. That volume tells me one thing: developers are tired of paying a heavy API tax just to filter user uploads. AI is just statistical probability. Run it locally.
#FALCONSAI
New blog post analyzing the top 50 entities with the most downloaded models on @huggingface 🤗!
The purpose here is to get an idea of the profile of the models with the greatest impact in open source (we are not interested in closed models here!).
Some key findings:
@sundarpichai , Quick idea, Allow funding exceptions to the Google Startups, that in leiu of funding if the company is pre-seed with X downloads from App store, HuggingFace etc. They can qualify for up to $500k in cloud credits. This actually helps Start-ups that need to scale.