Pedro Cuenca

Career

Pedro Cuenca is a machine learning engineer at Hugging Face whose work, according to the author bio in Hands-On Generative AI with Transformers and Diffusion Models, centers on diffusion software, models and applications. Before Hugging Face, Cuenca was co-founder and CTO of LateNiteSoft, working on the technology behind the Camera+ iPhone app, which used custom machine learning models for photography enhancement.¹

The same bio notes that Cuenca created deep-learning models for photography enhancement and super-resolution and was involved in the development and operations of DALL·E mini, whose README lists Cuenca among its authors.¹⁶

Open-source work

Several of Cuenca's Hugging Face projects bring models to Apple devices. Cuenca wrote the December 2022 post on using Stable Diffusion with Core ML on Apple Silicon, co-wrote the February 2023 post introducing the open-source Diffusers app for Mac, and in August 2023 introduced swift-transformers, an Apache-2.0 Swift package for running language models on device.⁷⁵⁸⁴⁹

Cuenca is listed among the authors in the citation for Hugging Face's Diffusers library and co-authored the Hugging Face launch posts for open models including Llama 2, Gemma and Llama 3. In June 2026 Cuenca co-wrote the second part of a Hugging Face series on profiling in PyTorch, which walks from nn.Linear to a fused MLP.^31011122

Writing

Cuenca co-wrote the book Hands-On Generative AI with Transformers and Diffusion Models, whose companion repository lists Cuenca among the authors alongside Apolinario Passos and others.¹

Work and contributions

• Diffusers Open-source code

  Listed among the authors in the citation entry of the Diffusers README and its CITATION.cff file; the library is released under the Apache 2.0 license.³

• swift-transformers Open-source code

  Author of the 2023 Hugging Face blog post releasing swift-transformers, a Swift package with tokenizers, Hub wrappers and generation algorithms for running language models on Apple devices through Core ML; the repository is under the Apache 2.0 license.⁴⁹

• Diffusers for Mac (swift-coreml-diffusers) Open-source code

  Co-author of the 2023 Hugging Face blog post on version 1.1 of the native Mac app, which runs text-to-image models converted to Core ML on the user's own machine.⁵⁸

• DALL·E mini Open-source code

  Listed in the README's author list and citation entry for the Apache-2.0 DALL·E mini project.⁶

Timeline

1. Dec 1, 2022Published the Hugging Face post on using Stable Diffusion with Core ML on Apple Silicon.⁷
2. Feb 24, 2023Co-wrote the post introducing Swift Diffusers, the open-source Diffusers app for Mac.⁵
3. Jul 18, 2023Co-authored Hugging Face's Llama 2 launch post.¹⁰
4. Aug 8, 2023Released swift-transformers for running language models on Apple devices.⁴
5. Feb 21, 2024Co-authored Hugging Face's post welcoming Google's Gemma open models.¹¹
6. Apr 18, 2024Co-authored Hugging Face's post welcoming Meta's Llama 3 models.¹²
7. Jun 11, 2026Co-wrote "Profiling in PyTorch (Part 2): From nn.Linear to a Fused MLP" on the Hugging Face blog.²

Other documented roles

Roles the sources document that are not confirmed as current.

• Co-founder and CTO, LateNiteSoft¹

Where they work

• Hugging Face 34 people in the wiki

Related people

• Sigbjørn SkjæretAlso at Hugging Face
• Adrien GallouëtAlso at Hugging Face
• Philipp SchmidAlso at Hugging Face
• Julien SimonAlso at Hugging Face
• Daniel BeveniusAlso at Hugging Face
• Peiyuan ZhangImage, video, and 3D generation · Inference and serving
• Piotr WilkinInference and serving · Developer tools
• Jeffrey MorganInference and serving · Developer tools
• Dane KnechtInference and serving · Developer tools
• Matthew PrinceInference and serving · Developer tools