datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
lucid-rag-ondevice
🎯 En bref
Ce dataset alimente le RAG embarqué de LUCID, une
application d'étude mobile. L'IA locale — de petits modèles quantifiés tournant directement
sur le téléphone — s'appuie sur ces chapitres pour ancrer ses quiz, son tutorat et ses
fiches sur le contenu réel des programmes, et non sur ses seules connaissances internes.
Pourquoi c'est différent — au lieu d'un modèle qui invente une réponse plausible,
LUCID récupère le passage de cours pertinent et génère à… See the full description on the dataset page: https://huggingface.co/datasets/clemNova/lucid-rag-ondevice.synthetic-medical-conversations-deepseek-v3
🍎 Synthetic Multipersona Doctor Patient Conversations.
Author: Nisten Tahiraj
License: MIT
🧠 Generated by DeepSeek V3 running in full BF16.
🛠️ Done in a way that includes induced errors/obfuscations by the AI patients and friendly rebutals and corrected diagnosis from the AI doctors. This makes the dataset very useful as both training data and retrival systems for reducing hallucinations and increasing the diagnosis quality.
🐧 Conversations… See the full description on the dataset page: https://huggingface.co/datasets/OnDeviceMedNotes/synthetic-medical-conversations-deepseek-v3.healthbench
THE CODE IS CURRENTLY BROKEN BUT THE DATASET IS GOOD!!
HealthBench Implementation for using Opensource Judges
Easy-to-use implementation of OpenAI's HealthBench evaluation benchmark with support for any OpenAI API-compatible model as both the system under test and the judge.
Developed by: Nisten Tahiraj / OnDeviceMednotes
License: MIT
Paper: HealthBench: Evaluating Large Language Models Towards Improved Human Health
Overview
This repository contains tools… See the full description on the dataset page: https://huggingface.co/datasets/OnDeviceMedNotes/healthbench.nih-rxnorm-dec-2-2024Source: https://datadiscovery.nlm.nih.gov/Drugs-and-Chemicals/RxNorm/t84b-g5db/about_data
Courtesy of the U.S. National Library of Medicine
on-device-face-liveness-detection
Mobile Face Liveness Detection
The dataset consists of videos featuring individuals wearing various types of masks. Videos are recorded under different lighting conditions and with different attributes (glasses, masks, hats, hoods, wigs, and mustaches for men).
The dataset is created on the basis of iBeta Level 1 Dataset
In the dataset, there are 4 types of videos filmed on mobile devices:
2D mask with holes for eyes - demonstration of an attack with a paper/cardboard… See the full description on the dataset page: https://huggingface.co/datasets/UniqueData/on-device-face-liveness-detection.tulu-v2-sft-mixturehh-rlhf-h4SlimOrcaoasst1_top1_2023-08-25health-canada-drug-products-database#Public Health Canada
##Drug Products Database Dump
This is publicly accessible open data uploaded here for easy of use.
Files can also directly be downloaded from here:
https://health-products.canada.ca/api/drug/
Updated as of January 5 2025
on-device-latency
On-Device Latency Benchmark
Real-world inference latency data for mobile-optimized LLMs, measured on actual phone hardware.
Hardware
Spec
Value
Device
Samsung S20 FE 5G
SoC
Snapdragon 865
RAM
8GB
OS
Android 13
Runtime
llama.cpp (4 threads)
Metrics
tokens_per_sec — Generation speed during inference
latency_ms_per_token — Time per generated token
ram_usage_mb — Peak RAM during inference
file_size_mb — GGUF model file size… See the full description on the dataset page: https://huggingface.co/datasets/dispatchAI/on-device-latency.Federated_Meta-learning_on_wearable_devices