Thomas Liang PRO

thliang01's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted an article 1 day ago

view article

Article

How I contributed a new model to the Transformers library using Codex

• 

3 days ago

•

37

upvoted an article 3 days ago

view article

Article

Building the Open Agent Ecosystem Together: Introducing OpenEnv

• 
• 
• +8

Oct 23, 2025

•

156

New activity in twinkle-ai/gemma-3-4B-T1-it 3 days ago

BFCL 相關事項詢問

#1 opened 29 days ago by

AaronWu901225

updated 4 datasets 4 days ago

twinkle-ai/finevision-zhtw

Updated 4 days ago • 16 • 2

twinkle-ai/finevoices-zhtw

Updated 4 days ago • 22 • 1

twinkle-ai/finetranslations-zhtw

Updated 4 days ago • 15 • 2

twinkle-ai/finepdfs-zhtw

Updated 4 days ago • 22 • 5

liked a Space 6 days ago

Sleeping

1

Fine Vision Album

📸

1

Collect Taiwan daily life photos for zhtw dataset

liked a dataset 16 days ago

HuggingFaceFW/fineweb-2

Viewer • Updated Oct 27, 2025 • 4.48B • 38.4k • 775

upvoted a paper 17 days ago

Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters

Paper • 2602.10604 • Published Feb 11 • 194

liked a model 20 days ago

twinkle-ai/twinkle-sqlcoder

Text Generation • 24B • Updated 20 days ago • 76 • 2

updated a collection 21 days ago

📋 Twinkle Eval Logs

Collection

Benchmark log generated with Twinkle Eval, recording the model's outputs for each prompt, see more in https://github.com/ai-twinkle/Eval • 22 items • Updated 9 days ago • 1

liked a model 21 days ago

nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16

Text Generation • 124B • Updated 9 days ago • 217k • 317

upvoted a collection 21 days ago

NVIDIA Nemotron v3

Collection

Open, Production-ready Enterprise Models • 15 items • Updated 2 days ago • 258

upvoted an article 21 days ago

view article

Article

Keep the Tokens Flowing: Lessons from 16 Open-Source RL Libraries

• 
• 
• +7

24 days ago

•

99

upvoted a collection 22 days ago

📋 Twinkle Eval Logs

Collection

Benchmark log generated with Twinkle Eval, recording the model's outputs for each prompt, see more in https://github.com/ai-twinkle/Eval • 22 items • Updated 9 days ago • 1

liked a Space 24 days ago

Running

2

Twinkle Evaluator Leaderboard

📊

2

Official Benchmark Leaderboard for Twinkle Eval

upvoted 2 collections 25 days ago

LLM PlayBooks

Collection

All useful playbooks for training LLM • 6 items • Updated 25 days ago • 2

🤏 Smol-Data

Collection

Tried and tested mixes for strong pretraining. Inspired by https://huggingface.co/blog/codelion/optimal-dataset-mixing • 14 items • Updated Mar 2 • 12

liked a Space 25 days ago

Running on CPU Upgrade

214

The Synthetic Data Playbook: Generating Trillions of the Finest Tokens

📝

214

Explore synthetic data experiments as an interactive bookshelf