Younes B PRO

ybelkada

·

AI & ML interests

Large Language Models, Quantization, Vision, Multimodality, Diffusion models

Recent Activity

updated a bucket about 4 hours ago

ybelkada/async-grpo-gsm8k-bucket

published a bucket 1 day ago

ybelkada/async-grpo-harbor-static-ca55b6-bucket

updated a bucket 1 day ago

ybelkada/elastic-inference

View all activity

Organizations

upvoted a collection 1 day ago

LFM2.5-24B Harness adapters

A collection of models trained with RL on specific adapters for more efficient agentic development • 4 items • Updated 2 days ago • 16

upvoted a collection 5 days ago

📚 Research Literature

Research on agent swarms, collective dynamics, and learning to coordinate. • 6 items • Updated 5 days ago • 2

upvoted an article 6 days ago

Article

Reverse Engineering Anything using small open models

ybelkada

•

6 days ago

• 5

upvoted 2 articles 20 days ago

Article

Falcon OCR Arabic: 270M Parameters State-of-the-Art Arabic OCR

tiiuae

•

26 days ago

• 36

Article

New in llama.cpp: Decision Models

ggml-org

•

30 days ago

• 107

upvoted a collection 26 days ago

SmolDataEnvs Multi-harness RL

Try the three environments, follow the tutorial, and explore the datasets, models and historical RL/SFT results. • 22 items • Updated 25 days ago • 4

upvoted a collection 6 months ago

Falcon-E BitNet

A collection of fine-tuned Falcon-E BitNet models using Axolotl. https://huggingface.co/blog/axolotl-ai-co/finetuning-ternary-llms-tii-axolotl • 7 items • Updated Apr 30 • 3

upvoted an article 6 months ago

Article

Training low-bit ternary models with Axolotl

axolotl-ai-co

•

Apr 30

• 9

upvoted a paper 7 months ago

Falcon Perception

Paper • 2603.27365 • Published Mar 28 • 17

upvoted an article 7 months ago

Article

Falcon Perception

tiiuae

•

Apr 1

• 71

upvoted 2 collections 7 months ago

Falcon Perception

Falcon-Perception and Falcon-OCR model: early-fusion, natively multimodal, dense Autoregressive Transformer models. • 5 items • Updated Aug 13 • 14

SigLino: Vision Foundation Models (SigLIP2 + DINOv3)

Vision encoders distilled from DINOv3 and SigLIP2 (MoE & Dense), and the OpenLVD-200M training data. CVPR 2026. • 7 items • Updated 10 days ago • 17

upvoted a collection 9 months ago

MioTTS

12 items • Updated Feb 14 • 11

upvoted a paper 9 months ago

SERA: Soft-Verified Efficient Repository Agents

Paper • 2601.20789 • Published Jan 28 • 13

upvoted a collection 9 months ago

Open Coding Agents

13 items • Updated Mar 5 • 56

upvoted an article 9 months ago

Article

Alyah ⭐️: Toward Robust Evaluation of Emirati Dialect Capabilities in Arabic LLMs

tiiuae

•

Jan 27

• 26

upvoted 2 collections 10 months ago

Falcon-H1-Tiny

A series of extremely small, yet powerful language models redefining capabilities at small scale • 19 items • Updated Aug 13 • 38

Falcon-H1R

5 items • Updated Aug 13 • 27

upvoted 2 papers 10 months ago

Falcon-H1R: Pushing the Reasoning Frontiers with a Hybrid Model for Efficient Test-Time Scaling

Paper • 2601.02346 • Published Jan 5 • 28

Learnable Multipliers: Freeing the Scale of Language Model Matrix Layers

Paper • 2601.04890 • Published Jan 8 • 45