QVAC

@qvac

Infinite intelligence. Local. Any Hardware. Peer-to-Peer Hyper Swarm. No cloud. No compromise. QVAC is the decentralized AI platform for humans and machines.

qvac.tether.io

Joined April 2025

Following

11.9K

Followers

Replies Reposts Media Articles

QVAC
@qvac
17h
Translation does not need a giant model. It needs one small enough to live on a phone and keep working offline. That is what TranslatePsy models are for, and they just shipped as two families: - AfriSLM across 19 Sub-Saharan African languages, - Nano, down to tens of megabytes
On-device translation: Why smaller, specialized AI models win
From networkworld.com
4
QVAC
@qvac
Sep 4
Hiraia is being built with the support of a Tether grant. The people who need AI most are the ones priced out of it. A subscription in a currency you do not earn, behind a connection you do not have, is not access. We fund work that removes those barriers, because intelligence
QVAC
@qvac
Sep 4
Hiraia.org is an AI science tutor for Filipino students created by @helloluis that is built with the QVAC SDK. It holds 50,000 science facts, 30,000 illustrations and 20,000 mini-quizzes. It can answer questions in Tagalog and English. Hiraia follows the Department
QVAC
@qvac
Sep 4
Hiraia.org is an AI science tutor for Filipino students created by @helloluis that is built with the QVAC SDK. It holds 50,000 science facts, 30,000 illustrations and 20,000 mini-quizzes. It can answer questions in Tagalog and English. Hiraia follows the Department
4
QVAC
@qvac
Sep 3
Introducing TranslatePsy-EuroNano and TranslatePsy-AfriNano, two new additions to the TranslatePsy family: 9 European languages and 8 African languages, in models small enough to run in a browser. One multilingual checkpoint replaces a folder of bilingual models. Model cards:
9
QVAC
@qvac
Sep 2
There are two main reasons for the limitations of African language machine translation: the training data and the expensive compute requirements of frontier open-models. In our latest work, we are addressing both by shipping compact small translation language models as well as
QVAC
@qvac
Sep 2
Introducing TranslatePsy-AfriSLM: state-of-the-art machine translation for 19 Sub-Saharan African languages using models small enough to run on a laptop or a smartphone. Our smallest LM, at just 0.8B parameters, matches or beats systems up to 152 times its size, including
00:00
2