🔄 In a Training Loop

swappy

rycerzes

·

https://rycerz.es

AI & ML interests

Deep learning, LLMs/VLMs, RL

Recent Activity

updated a dataset 6 days ago

rycerzes/atarashi-dataset

published a dataset 6 days ago

rycerzes/atarashi-dataset

upvoted an article 7 days ago

Training a coding agent using the OpenCode harness in remote HF sandboxes with TRL and OpenEnv

View all activity

Organizations

liked a Space 21 days ago

Hugging Trace Viewer

Browse and share Perfetto traces from Hugging Face

liked a model 24 days ago

rycerzes/nirjas-gate

32.3M • Updated Jul 13 • 7 • 1

liked a Space about 2 months ago

Encoder-Free VLM

Train Your Own Encoder-Free VLM in $100

liked 2 Spaces 3 months ago

Token-In, Token-Out Done Right

Explore an interactive simulation while reading the article

Bringing paper to life: A modern template for scientific writing

Explore an interactive galaxy visualization of scientific article

liked 3 models 3 months ago

facebook/VGGT-Omega

Updated May 14 • 234

Qwen/WebWorld-8B

Text Generation • 8B • Updated May 8 • 737 • 65

Zyphra/ZAYA1-8B

9B • Updated Jun 26 • 32.7k • 588

liked a Space 3 months ago

The ultimate guide to RL environments: building and scaling them in the LLM era

Building and scaling RL environments for LLM training

liked 5 models 4 months ago

LiquidAI/LFM2.5-Audio-1.5B

Audio-to-Audio • 1B • Updated Mar 30 • 1.28k • 453

IDEA-Research/Rex-Omni

Image-Text-to-Text • 4B • Updated Oct 16, 2025 • 22.1k • 62

google/gemma-4-E2B-it

Any-to-Any • 5B • Updated 26 days ago • 3.94M • 895

facebook/sam3

Mask Generation • 0.9B • Updated Nov 20, 2025 • 2.29M • 2.65k

openbmb/MiniCPM-o-4_5

Any-to-Any • 9B • Updated 12 days ago • 967k • 1.46k

liked a dataset 4 months ago

markov-ai/computer-use

Viewer • Updated Feb 13 • 313 • 505 • 69

liked a Space 5 months ago

The Synthetic Data Playbook: Generating Trillions of the Finest Tokens

Visualize synthetic‑data experiments as an interactive bookshelf

liked a Space 9 months ago

The Smol Training Playbook

The secrets to building world-class LLMs

liked a Space 10 months ago

FineVision: Open Data is All You Need

A new open-source dataset for training VLMs