🏗️ Building on HF

Parag Ekbote

AINovice2005

13 17 25

clem's profile picture

thliang01's profile picture

DollarSign's profile picture

https://paragekbote.github.io/

ParagEkbote
parag-ekbote

AI & ML interests

ML Engineer passionate about taking models from research to production. 1 year supporting tech startups. Active OSS contributor.

Recent Activity

upvoted an article 1 day ago

Prediction preview images with modular diffusers

upvoted an article 1 day ago

Using masks with LoRAs in Modular Diffusers

updated a Space 2 days ago

AINovice2005/carbon-enrichment-pipeline-diagram-2

View all activity

Organizations

AINovice2005 's collections 6

Carbon Genomic Corpus Enrichment Datasets

A collection of enriched datasets derived from Carbon genomic pretraining corpus. The artifacts support biological analysis and comparative studies.

AINovice2005/carbon-embeddings

Viewer • Updated 11 days ago • 251k • 394
AINovice2005/carbon-cpu-enriched-sequences

Viewer • Updated 11 days ago • 32.4M • 694
AINovice2005/carbon-likelihood-stats

Viewer • Updated 11 days ago • 251k • 609
AINovice2005/carbon-cpu-enriched-sequences-sampled

Viewer • Updated about 1 month ago • 3.22M • 659

SmolLM-Smashed: Tiny Giants, Optimized for Speed

SmolLM-Smashed is a collection of optimized language models. Each model is quantized and compiled for maximum efficiency while preserving performance.

AINovice2005/SmolLM3-3B-smashed

Text Generation • Updated Oct 4, 2025 • 67
AINovice2005/SmolLM2-1.7B-smashed

Text Generation • Updated Oct 4, 2025 • 14
AINovice2005/SmolLM2-360M-smashed

Text Generation • Updated Oct 4, 2025 • 13
AINovice2005/SmolLM-360M-smashed

Text Generation • Updated Oct 4, 2025 • 17

🚀 Optimized Models: torchao & Pruna Quantization

Quantized Models using torchao & Pruna for efficient inference and deployment.

AINovice2005/quantized-GLM-4.1V-9B-Thinking

Image-Text-to-Text • Updated Jul 17, 2025 • 16
AINovice2005/quantized-SmolVLM2-2.2B-Base

Image-Text-to-Text • Updated Jul 17, 2025 • 9
AINovice2005/Sana_600M_ControlNet_HED-smashed

Text-to-Image • Updated Jul 21, 2025 • 5
AINovice2005/quantized-SmolLM3-3B

Text Generation • Updated Jul 17, 2025 • 40 • 1

CICFlow-Meter ModernBERT LoRA

A collection of ModernBERT‑based binary classifiers fine‑tuned with LoRA adapters at ranks 4, 8, and 16 for efficient network flow analysis.

AINovice2005/ModernBERT-base-lora-cicflow-1m-r4

Fill-Mask • Updated May 11 • 29 • 1
AINovice2005/ModernBERT-base-lora-cicflow-1m-r8

Fill-Mask • Updated Mar 18 • 22
AINovice2005/ModernBERT-base-lora-cicflow-1m-r16

Fill-Mask • Updated Mar 18 • 18

👨‍🍳Cooking with HF: HPO with Transformers and Optuna

My HF recipe space and model.

AINovice2005/bert-imdb-optuna-hpo

Fill-Mask • 0.1B • Updated Jul 9, 2025 • 10
Sleeping

Hf Cookbook Hpo

📊

Optuna-based hyperparameter search for transformer models

Blazing Kernels

AINovice2005/attention-int8

Updated Mar 7 • 2