Ahmet's picture

Building on HF

Ahmet

atasoglu

·

https://atasoglu.github.io/

AI & ML interests

NLP, LLMs.

Recent Activity

liked a Space 13 minutes ago

nvidia/LocateAnything

liked a model about 10 hours ago

unsloth/gemma-4-E4B-it-qat-mobile-GGUF

upvoted a collection about 10 hours ago

View all activity

Organizations

upvoted a collection about 10 hours ago

Gemma 4 QAT

Gemma 4 QAT (Quantization-Aware Training) for 3x less memory use and near original accuracy. • 16 items • Updated about 15 hours ago • 26

upvoted 2 collections about 11 hours ago

Gemma 4 QAT Mobile

4 items • Updated about 15 hours ago • 17

Gemma 4 QAT Q4_0

19 items • Updated about 11 hours ago • 46

upvoted an article 18 days ago

Article

Introducing ColQwen-Omni: Retrieve in every modality

manu

•

Jul 17, 2025

• 77

upvoted a collection 24 days ago

jina-embeddings-v5-omni

Multimodal (text + image + video + audio) embedding models aligned with jina-embeddings-v5-text-*. Two sizes, four task variants each. • 27 items • Updated 25 days ago • 36

upvoted a collection 28 days ago

privacy-filter

OpenAI's privacy-filter fine0tuned models • 6 items • Updated about 1 month ago • 10

upvoted 5 collections about 1 month ago

Gemma-4 Assistant (MTP)

4 items • Updated May 5 • 27

Mistral Medium 3.5

Our first flaship models handling instruction-following, reasoning, and coding in a single set of opened-weights. • 2 items • Updated Apr 29 • 17

Privacy Filter

3 items • Updated Apr 22 • 1

Qwen3.6

4 items • Updated Apr 22 • 396

gpt-oss-safeguard

gpt-oss-safeguard-120b and gpt-oss-safeguard-20b are safety reasoning models built-upon gpt-oss • 2 items • Updated Oct 29, 2025 • 74

upvoted 2 collections 2 months ago

Gemma 4

Gemma 4 is Google's new model family including including E2B, E4B, 26B-A4B, and 31B. • 36 items • Updated about 17 hours ago • 209

Gemma 4

15 items • Updated 3 days ago • 914

upvoted an article 2 months ago

Article

Tokenization is Killing our Multilingual LLM Dream

omarkamali

•

Mar 15

• 12

upvoted 4 collections 3 months ago

Holo2

Holo2 - Cost-Efficient Models for Cross-Platform Computer-Use Agents • 4 items • Updated Feb 2 • 27

Inference Optimized Checkpoints (with Model Optimizer)

A collection of generative models quantized and optimized for inference with Model Optimizer. • 75 items • Updated about 8 hours ago • 166

ERDES — Ocular Ultrasound Classification All Tasks

ERDES benchmark: 8 architectures × 5 ocular ultrasound classification tasks. Paper: arXiv:2508.04735 | Dataset: pcvlab/erdes • 59 items • Updated Mar 5 • 2

magibu-q-series

5 items • Updated Mar 6 • 3

upvoted a paper 3 months ago

ERDES: A Benchmark Video Dataset for Retinal Detachment and Macular Status Classification in Ocular Ultrasound

Paper • 2508.04735 • Published Aug 5, 2025 • 2

upvoted a collection 3 months ago

Qwen3.5

21 items • Updated Mar 9 • 1.67k