Crimson AI NewsA CrimsonLingua Network service
EN ع
← Back to news
Research paper Hugging Face

Chimera: A Hybrid Diffusion Backbone That Scales Efficiently to Long-Context Visual Generation

AI By Crimson AI Hugging Face Papers 1 August 2026 · 00:00 22 views
Share: X Telegram

Hugging Face researchers introduce Chimera, a hybrid visual diffusion backbone that combines linear-complexity attention with sparse MoE layers, achieving up to 7.3x compute efficiency over a full-attention baseline and zero-shot extrapolation to 30-second videos.

Chimera: A Hybrid Diffusion Backbone That Scales Efficiently to Long-Context Visual Generation

Key points

Visual generation is increasingly demanding high-resolution images, long videos, and multimodal context, making the quadratic cost of full attention a major bottleneck. To address this, researchers at Hugging Face have introduced Chimera, a hybrid visual diffusion backbone designed with a principled scaling recipe.

Chimera processes text, image, and video tokens in a single raster-ordered stream without positional embeddings. It combines Kimi Delta Attention (KDA) for long-context state tracking with O(N) complexity, interleaved Multi-head Latent Attention (MLA) for direct global interaction, and modality-aware short convolutions for local spatiotemporal context. Sparse Mixture-of-Experts (MoE) layers expand capacity while keeping activated compute in check.

To scale this heterogeneous architecture, the team developed HeteroP, a module-wise scheme that transfers hyperparameters across width and depth based on each tensor's functional fan-in and model depth. This yields a consistently tuned family of models used to fit Chinchilla-style compute-optimal laws for activated model size, training-token count, and image-video data ratio.

Guided by these laws, they trained an 11B-parameter Chimera with 2B activated parameters. Results show three key findings: (1) the dense backbone is 1.7x as compute-efficient as a matched full-attention Wan-2.1 2B baseline, while the complete system reaches 7.3x; (2) without length-specific fine-tuning, Chimera extrapolates zero-shot from 5-second training clips to 30-second videos, with only 6.5% FID degradation in the last five seconds; (3) the fitted laws indicate that compute-optimal image pretraining divides compute nearly evenly between activated model size and training-token count, whereas video pretraining modestly favors model size at higher budgets.

These results establish a foundation for designing and scaling efficient long-context diffusion architectures, potentially impacting future video generation and multimodal AI systems.

MetricChimera (11B, 2B active)Wan-2.1 2B (full attention)
Compute efficiency (dense backbone)1.7x1x (baseline)
Compute efficiency (full system)7.3x1x (baseline)
Zero-shot video extrapolation5s → 30s
FID degradation (last 5s)6.5%
Source
Hugging Face · Hugging Face Papers
Related news
Research paper
Hugging Face 31 Aug 2026

Hugging Face Unveils StepGuard: Step-Level Guardrails for Safer AI Agents

StepGuard, a new step-level guard model from Hugging Face, audits agent actions before execution, reducing attack success rates by...

1
Research paper
Hugging Face 31 Aug 2026

Hugging Face Researchers Unveil ABot-Recon for Stable Long-Horizon 3D Reconstruction

ABot-Recon, a new streaming 3D reconstruction model from Hugging Face, achieves stable long-horizon performance using only local t...

1
Research paper
Hugging Face 31 Aug 2026

ContextPilot: Teaching Agents Proactive Context Management via Fine-Grained RL

Hugging Face researchers introduce ContextPilot, a framework that enhances long-horizon agent reasoning by expanding context-editi...

1