Younes Belkada
Listed author of Hugging Face's TRL library and co-author of the StarCoder2, Zephyr, and LLM.int8() papers
Younes Belkada is listed among the authors of TRL, Hugging Face's Apache-2.0 library for post-training transformer models, and is a co-author of the Zephyr and StarCoder2 papers. DeepLearning.AI's short course Quantization Fundamentals with Hugging Face names Belkada as an instructor and a Machine Learning Engineer at Hugging Face. Belkada is also a co-author of LLM.int8() (2022), which described 8-bit matrix multiplication that made inference with very large transformer models possible on a single server with consumer GPUs. Belkada's 2026 paper on learnable multipliers for language model matrix layers lists the Falcon LLM Team as Belkada's affiliation.123465
Work
TRL (Transformers Reinforcement Learning)Open-source code
Listed among the authors in TRL's citation file; TRL is an Apache-2.0 toolkit for post-training transformer models with SFT, reward modeling, DPO, and GRPO.1
StarCoder2Published research
Co-author of "StarCoder 2 and The Stack v2: The Next Generation" (2024).3
Zephyr (external site: arxiv.org)Published research
Co-author of "Zephyr: Direct Distillation of LM Alignment" (2023), on aligning a smaller chat model with distilled direct preference optimization.2
LLM.int8() (external site: arxiv.org)Published research
Co-author, with Tim Dettmers, Mike Lewis, and Luke Zettlemoyer, of "LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale" (2022); the abstract says the software is open source.6
Other documented roles
Roles the sources document but that are not confirmed as current.
- Machine Learning Engineer, Hugging Face4
This profile covers public professional work only, from the sources below. It is not an endorsement or a ranking. To correct or remove it, use the correction form.