🔄 In a Training Loop

Keylhan Paumard--André

keypa

5 34 311

keypaa

AI & ML interests

Efficient deep learning, LLM fine-tuning, inference optimization, model compression, distributed training, GPU systems, open-source AI infrastructure

Recent Activity

updated a dataset 1 day ago

keypa/vision-adapter-embeddings

updated a dataset 4 days ago

keypa/vision-adapter-images

published a dataset 5 days ago

keypa/vision-adapter-embeddings

View all activity

Organizations

upvoted a paper 22 days ago

Metis: Memory Foundation Model

Paper • 2607.26760 • Published 25 days ago • 272

upvoted a collection 23 days ago

QUEST

Collection

17 items • Updated Jul 1 • 26

upvoted an article 26 days ago

Article

Giving a 753B Model Eyes: Reproducing GLM-5.2 Vision on 4× 96 GB GPUs

0xSero

•

27 days ago

• 2

upvoted an article about 1 month ago

Article

Kimi K3 Model Overview: 2.8T Parameters, MXFP4 Quantization, and What the Open Weights Mean for the Community

ResterChed

•

Jul 17

• 197

upvoted 2 collections 2 months ago

Cerebras REAP

Collection

Sparse MoE models compressed using REAP (Router-weighted Expert Activation Pruning) method • 30 items • Updated Feb 25 • 152

Apodex-1

Collection

4 items • Updated Jun 8 • 36

upvoted 2 collections 3 months ago

Gemma 4 QAT Q4_0

Collection

19 items • Updated Jul 21 • 159

Qwen3

Collection

84 items • Updated Dec 31, 2025 • 1.85k

upvoted 2 articles 3 months ago

Article

Introducing the Ettin Reranker Family

tomaarsen

•

May 19

• 56

Article

From Zero to GPU: A Guide to Building and Scaling Production-Ready CUDA Kernels

drbh, danieldk

•

Aug 18, 2025

• 110

upvoted 2 collections 3 months ago

Qwen3.6

Collection

4 items • Updated Apr 22 • 490

Qwen3.5

Collection

21 items • Updated Mar 9 • 1.75k

upvoted a paper 3 months ago

WebWorld: A Large-Scale World Model for Web Agent Training

Paper • 2602.14721 • Published Feb 16 • 19

upvoted a paper 4 months ago

ProgramBench: Can Language Models Rebuild Programs From Scratch?

Paper • 2605.03546 • Published May 5 • 4

upvoted an article 4 months ago

Article

Introducing NVIDIA Nemotron 3 Nano Omni: Long-Context Multimodal Intelligence for Documents, Audio and Video Agents

nvidia

•

Apr 28

• 62

upvoted a paper 4 months ago

RAG-Anything: All-in-One RAG Framework

Paper • 2510.12323 • Published Oct 14, 2025 • 84

upvoted a collection 4 months ago

LeWM

Collection

Official checkpoints and datasets related to LeWM paper. • 9 items • Updated Mar 27 • 55

upvoted an article 4 months ago

Article

Building a Fast Multilingual OCR Model with Synthetic Data

nvidia

•

Apr 17

• 35

upvoted a paper 4 months ago

TREX: Automating LLM Fine-tuning via Agent-Driven Tree-based Exploration

Paper • 2604.14116 • Published Apr 15 • 13

upvoted a collection 4 months ago

Bonsai

Collection

1-bit Bonsai models • 7 items • Updated Jun 4 • 216

Keylhan Paumard--André

AI & ML interests

Recent Activity

Organizations

keypa's activity

Giving a 753B Model Eyes: Reproducing GLM-5.2 Vision on 4× 96 GB GPUs

Kimi K3 Model Overview: 2.8T Parameters, MXFP4 Quantization, and What the Open Weights Mean for the Community

Introducing the Ettin Reranker Family

From Zero to GPU: A Guide to Building and Scaling Production-Ready CUDA Kernels

Introducing NVIDIA Nemotron 3 Nano Omni: Long-Context Multimodal Intelligence for Documents, Audio and Video Agents

Building a Fast Multilingual OCR Model with Synthetic Data