🔄 In a Training Loop

Lewis Tunstall PRO

lewtun

huggingface

·

https://lewtun.github.io/blog/

AI & ML interests

LLMs, LLMs, LLMs

Recent Activity

upvoted an article about 19 hours ago

tokenizers v1: encode, decode and scaling, measured

View all activity

Organizations

upvoted an article about 19 hours ago

Article

tokenizers v1: encode, decode and scaling, measured

+2

ArthurZ, sbrandeis, mcpotato, lysandre

•

2 days ago

• 64

upvoted an article 9 days ago

Article

Async GRPO with LoRA across HF Jobs: a bucket, a proxy, and no NCCL

+2

aminediroHF, qgallouedec, kashif, sergiopaniego

•

13 days ago

• 50

upvoted a paper 19 days ago

AI Research Preference Models

Paper • 2608.13940 • Published 29 days ago • 11

upvoted an article 21 days ago

Article

Introducing @huggingface/kernels: 200+ WebGPU Kernels for Local AI

nico-martin, Xenova

•

22 days ago

• 79

upvoted a paper about 1 month ago

Training AI Scientists to Replicate Research

Paper • 2608.13331 • Published Aug 13 • 3

upvoted an article about 1 month ago

Article

State of Open Models: Summer 2026 Observations

+1

AdinaY, multimodalart, irenesolaiman

•

Aug 14

• 213

upvoted a paper about 1 month ago

Weak-to-Strong Generalization via Direct On-Policy Distillation

Paper • 2607.05394 • Published Jul 8 • 144

upvoted a collection about 1 month ago

Muse Glimmer

Muse Glimmer 30B: multimodal agentic model for local deployment. BF16 weights, GGUF k-quants, ExecuTorch builds, DFlash drafter. • 4 items • Updated Aug 10 • 109

upvoted 2 papers about 2 months ago

Understanding Reasoning from Pretraining to Post-Training

Paper • 2607.16097 • Published Jul 17 • 28

Kimi K3: Open Frontier Intelligence

Paper • 2607.24653 • Published Jul 27 • 523

upvoted an article about 2 months ago

Article

Anatomy of a Frontier Lab Agent Intrusion: A Technical Timeline of the July 2026 Incident

+2

hlarcher, XciD, raphael-gl, chris-rannou

•

Jul 27

• 504

upvoted 3 articles 2 months ago

Article

Security incident disclosure — July 2026

system

•

Jul 16

• 797

Article

Native-speed vLLM transformers modeling backend

hmellor, lysandre

•

Jul 8

• 76

Article

J-Space: Yet Another LLM Mind Reader?

dlouapre

•

Jul 13

• 43

upvoted a paper 3 months ago

AsyncOPD: How Stale Can On-Policy Distillation Be?

Paper • 2606.24143 • Published Jun 23 • 31

upvoted a changelog 3 months ago

Hugging Face Changelog

Share your feedback with us

Jun 26

• 139

upvoted 3 articles 3 months ago

Article

Harness, Scaffold, and the AI Agent Terms Worth Getting Right

sergiopaniego, ariG23498

•

May 25

• 148

Article

GLM-5.2: Built for Long-Horizon Tasks

zai-org

•

Jun 17

• 142

Article

PhysicsIntern: from an Autonomous Benchmark-runner to a Research Sidekick

dlouapre

•

Jun 11

• 7

upvoted an article 4 months ago

Article

Reachy Mini goes fully local

A-Mahla, andito

•

May 27

• 74