Steffen Röcker's picture

Steffen Röcker PRO

sroecker

·

https://x.com/sroecker

AI & ML interests

Local models

Recent Activity

liked a model 1 day ago

teapotai/teapotllm

upvoted a collection 3 days ago

liked a dataset 6 days ago

nvidia/Llama-Nemotron-Post-Training-Dataset-v1

View all activity

Organizations

sroecker's activity

upvoted a collection 3 days ago

Gemma 3 INT4

These are converted from the official QAT INT4 Flax checkpoints on Kaggle • 4 items • Updated 4 days ago • 1

upvoted a paper 8 days ago

SmolDocling: An ultra-compact vision-language model for end-to-end multi-modal document conversion

Paper • 2503.11576 • Published 11 days ago • 75

upvoted 2 collections 12 days ago

reranking series v2

V2 crispy rerank series • 2 items • Updated 13 days ago • 20

DeepHermes

Preview models of hybrid reasoner Hermes series • 6 items • Updated 12 days ago • 27

upvoted a collection 20 days ago

Q-Filters

Pre-computed Q-Filters for efficient KV cache compression. • 15 items • Updated 22 days ago • 6

upvoted a collection 25 days ago

Granite 3.2 Language Models

3 items • Updated 27 days ago • 19

upvoted a collection 26 days ago

DeepSeek-R1-Distill Quantized

18 items • Updated Feb 7 • 14

upvoted a collection 28 days ago

olmOCR

olmOCR is a document recognition pipeline for efficiently converting documents into plain text. olmocr.allenai.org • 4 items • Updated 6 days ago • 101

upvoted 2 collections about 1 month ago

SigLIP 2

OpenCLIP and timm SigLIP 2 models • 45 items • Updated Feb 21 • 14

ModernGLiClass

GLiClass with ModernBERT backbone • 4 items • Updated 18 days ago • 10

upvoted an article about 1 month ago

Article

DABStep: Data Agent Benchmark for Multi-step Reasoning

Feb 4

• 67

upvoted a paper about 1 month ago

On Teacher Hacking in Language Model Distillation

Paper • 2502.02671 • Published Feb 4 • 18

upvoted a collection about 2 months ago

EuroLLM

4 items • Updated Feb 21 • 30

upvoted a paper about 2 months ago

SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training

Paper • 2501.17161 • Published Jan 28 • 114

upvoted a collection about 2 months ago

R1 Multilingual

5 items • Updated Jan 31 • 10

upvoted a paper about 2 months ago

WILDCHAT-50M: A Deep Dive Into the Role of Synthetic Data in Post-Training

Paper • 2501.18511 • Published Jan 30 • 19

upvoted a collection about 2 months ago

Tulu 3 Models

All models released with Tulu 3 -- state of the art open post-training recipes. • 11 items • Updated 12 days ago • 96

upvoted an article about 2 months ago

Article

Open-R1: a fully open reproduction of DeepSeek-R1

Jan 28

• 823