Finally, they added a wide diffusion head the DiTDH variant. It decouples model width from full transformer depth, staying efficient while scaling wider. Result: 2.16 FID on ImageNet-256. RAE-DiTDH outperforms every VAE-based diffusion model at every scale.
OPEN SOURCE
-
Noise tweaks boost diffusion performance dramatically
By
–
Then came the magic: dimension-dependent noise scheduling and noise-augmented decoding. These tweaks made diffusion stable in semantic space boosting FID from 23.08 → 4.28. RAE models now converge 47× faster than SiT or REPA.
-
RAE + DINOv2-B beats VAEs with sharper images
By
–
Everyone assumed semantic encoders couldn’t reconstruct images.
Turns out, they can and better than VAEs. RAE + DINOv2-B achieves 0.49 rFID vs 0.62 for SD-VAE, with 6× less compute. That’s fewer FLOPs and sharper reconstructions. -
RAEs improve diffusion models over VAEs
By
–
Today, most diffusion models still use VAEs built on 2021 tech. They compress images into low-dimensional latents (like 4 channels). That’s why diffusion models lose global structure and texture fidelity. RAEs fix this by encoding rich semantic features directly from pretrained
-
New Paper Revolutionizes Diffusion Models
By
–
Holy shit…Diffusion just leveled up A new paper “Diffusion Transformers with Representation Autoencoders” basically kills the VAE era. Instead of the old VAE bottleneck, they use representation autoencoders (RAEs) built from pretrained encoders like DINO or SigLIP. The
-
Open datasets trend enables independent model training
By
–
I love the diversity of trending open datasets these days. There’s no excuse anymore not to train your own models! – Fineweb and a shuffle of it by @karpathy – Webscale-RL, a large-scale reinforcement learning dataset from @salesforce – SVQ, an audio dataset from @Google –
-
Open Agent Builder: Visual Workflow for AI Agents
By
–
Build, test, and deploy AI agent workflows with a visual no-code interface! Open Agent Builder is a visual workflow builder for creating complex AI agent workflows using a drag-and-drop interface 100% Open Source.
-
Open Source LLM Inference: Current State of Chaos
By
–
this is what opensource LLM inference
looks like in my head > pure chaos
> lousy & misplaced integrations > everything half-broken, somehow still runs we're so early, and there's a lot of work to do -
Open Source AI Week in San Francisco Live Coverage
By
–
Follow our live blog as we head to Open Source AI Week in San Francisco:
-
NVIDIA’s Open Source Progress Finally Getting Recognition
By
–
Great to see @NVIDIAAI
's recent great progress on open source starting to get noticed. I feel like they been really under the radar. Perhaps because their earlier models were buried under crappy licenses. But they've really come around in recent months.