Video yükleniyor...
Video Yüklenemedi
We recently open sourced a background removal benchmark where votes rank state-of-the-art model. We knew internally Photoroom had the best technology in the world but we needed to prove it. Proving it means good data and trust in the data. This trust comes from a fully transparent and open... show more
27,697 görüntüleme • 1 yıl önce •via X (Twitter)
11 Yorum

Here is the full process of the benchmark

Our speech-to-text models are the most accurate on the market with top rankings across industry benchmarks. - The highest accuracy rates—up to 95% - Up to 30% fewer hallucinations than other leaders - Low latency—63 minutes converts in 35 seconds Try via API for free today 👇

By far the best! Nothing is more annoying than going on Canva, trying to remove the background, and it didn't work. And then trying to do it manually, and that's even worse.

can confirm it’s the best. soon available on replicate? :) @fofrAI @replicate

love love photoroom! only request is to improve bg removal for photos taken inside a car 🫠

Do you have an example ?

@ylecun That's an impressive initiative, great way to contribute to the open-source community.

Looks really good. Would check it out

@ylecun @ylecun, it's exciting to see developments like these in open-source projects! Collaborative efforts foster innovation and transparency in AI. How might this impact future model evaluations? 🤔 #OpenSourceInnovation

@ylecun This collaborative move toward transparency enhances our innovation culture. How can we leverage open-source more effectively? 🤔 #CollaborationMatters

Impressive transparency and trust-building process! Great work.
