bagel.com (@bageldotcom) / X

bagel.com

654 posts

bagel.com

@bageldotcom

Diffusion Models. We're hiring: bagel.com/careers

Joined June 2023

Pinned
bagel.com
@bageldotcom
May 28
Today we're releasing Paris 2.0, to our knowledge the first decentralized-trained video generation model. At Bagel Labs, we believe frontier models should not require homogeneous clusters of premium, supply constrained GPUs. Paris 1.0 proved this for image generation. Paris 2.0
00:42
bidhan
@bidhan
May 28
We're releasing Paris 2.0, which, to our knowledge, is the world's first decentralized trained video generation model. We benchmarked it against a monolithic model trained on the same data and compute budget, and Paris 2.0 outperformed the monolithic by ~2x on FVD benchmark.
4.5K
bagel.com reposted
bidhan
@bidhan
May 28
We're releasing Paris 2.0, which, to our knowledge, is the world's first decentralized trained video generation model. We benchmarked it against a monolithic model trained on the same data and compute budget, and Paris 2.0 outperformed the monolithic by ~2x on FVD benchmark.
00:00
447K
bagel.com
@bageldotcom
Apr 27
3.2K
bagel.com
@bageldotcom
Mar 15
Replying to @bageldotcom
NVIDIA GTC - Generative World Models for Startups | by Bagel Labs & Alumni Ventures · Luma
From luma.com
1.2K
bagel.com
@bageldotcom
Mar 15
In town for NVIDIA GTC? If you're building generative world models or investing in the people who are - we're putting the right people in one room for you tomorrow night in Palo Alto. Co-hosted by Alumni Ventures. Signup link below.
00:00
11K
bagel.com reposted
Gin Jiang @CVPR
@ZhiyingJ
Mar 11
we managed to extend DDM to heterogeneous objectives! one step closer to decentralized AI XD tldr: experts trained in complete isolation, with different objectives, no communication - and mixing them beats making them all train the same way. 20-48G memory per expert
bidhan
@bidhan
Mar 11
Excited to share that Bagel Labs' paper got accepted at CVPR 2026. A lot of the most important diffusion model research has historically stayed inside frontier labs. We're bringing more of that in the open through open science and open infrastructure. In this work we showcase
2.5K
bagel.com reposted
bidhan
@bidhan
Mar 11
Excited to share that Bagel Labs' paper got accepted at CVPR 2026. A lot of the most important diffusion model research has historically stayed inside frontier labs. We're bringing more of that in the open through open science and open infrastructure. In this work we showcase
4.9K
bagel.com
@bageldotcom
Mar 8
Diffusion models are becoming the foundation for image, video, and world models. We are hosting a founders and investors gathering on that topic during NVIDIA GTC week, co-hosted by our friends at Alumni Ventures. Mar 16, Menlo Park. Sign up below. luma.com/nvidia-gtc-gen…
5.8K
bagel.com reposted
bidhan
@bidhan
Feb 6
Being at the frontier - by the definition of it - means creating the frontier. You don't get to be at the frontier by following someone else. And creating the frontier often means discoveries that go against the established knowledge. We recently made such a discovery about
52K
bagel.com reposted
Deep-ML
@real_deep_ml
Jan 22
Can't wait for the live stream! in honor of it we just added a new question and interactive playground based on the paper
00:00
Yacine Mahdid
@yacinelearning
Jan 21
alright folks tomorrow january 22 from 10h-12h AM EST we're going to dive into decentralized diffusion models by reviewing the paris model from bagel labs I even managed to lock in @bidhan for an interview on why how what that's a good thing to even do tune in!
7.2K
bagel.com reposted
Yacine Mahdid
@yacinelearning
Jan 21
alright folks tomorrow january 22 from 10h-12h AM EST we're going to dive into decentralized diffusion models by reviewing the paris model from bagel labs I even managed to lock in @bidhan for an interview on why how what that's a good thing to even do tune in!
Yacine Mahdid
@yacinelearning
Jan 10
this weekend I'll be diving deep into decentralized training for diffusion models with the paris model and the bagel team (what a sentence)
27K
bagel.com reposted
mirian
@mirimayer
Jan 15
no better way to celebrate bagel day 🥯
2.2K
bagel.com reposted
bidhan
@bidhan
Dec 17, 2025
NeurIPS takeways (better late than never) 1. real AGI needs real continual learning - models that can keep learning without catastrophic forgetting. 2. model architectures need to be "stateful" for building accurate world models for games and robotics. 3. diffusion models are
7.4K