Hugging Face's logo

Collections

Discover the best community collections!

Collections trending this week
MultiVectorEncoder integrations
Collection by
Oct 8
Local Image Pipelines
What I load in ComfyUI, and which models do true instruction edits vs. needing img2img. Includes the encoders, VAEs and ControlNets you need.
Serving Economics
Inference cost decides your architecture. These papers, read top to bottom, build the mental model: where the FLOPs and the bytes actually go.
One GPU, Real Work
Everything I run on a 64GiB unified-memory Strix Halo box, with my own measured numbers. The finding: MoE decodes ~6x faster than dense here.
Benin · Country Data Coverage
194 official ML-ready Benin open-data resources, plus related Electric Sheep Africa datasets.
MultiVectorEncoder integrations
Collection by
Oct 8
Local Image Pipelines
What I load in ComfyUI, and which models do true instruction edits vs. needing img2img. Includes the encoders, VAEs and ControlNets you need.
One GPU, Real Work
Everything I run on a 64GiB unified-memory Strix Halo box, with my own measured numbers. The finding: MoE decodes ~6x faster than dense here.
Serving Economics
Inference cost decides your architecture. These papers, read top to bottom, build the mental model: where the FLOPs and the bytes actually go.
Benin · Country Data Coverage
194 official ML-ready Benin open-data resources, plus related Electric Sheep Africa datasets.