QVAC TranscribePsy: State-of-the-Art On-Device Multilingual Speech Recognition
• 7
AI & ML interests
None defined yet.
Recent Activity
View all activity
Articles
Compact multilingual Automatic Speech Recognition (ASR) model that transcribes speech to text, and purpose-built to run on the phone in your pocket.
-
qvac/TranscribePsy-250M-EU-sub-gguf
Automatic Speech Recognition • 0.3B • Updated • 232 • 3 -
qvac/TranscribePsy-250M-EU
Automatic Speech Recognition • 0.3B • Updated • 73 • 4 -
qvac/TranscribePsy-250M-EU-sub-coreml
Automatic Speech Recognition • 14.6M • Updated • 44 • 2 -
qvac/TranscribePsy-250M-EU-sub
Automatic Speech Recognition • 0.3B • Updated • 34 • 2
Compact multilingual sequence to sequence neural machine translation (NMT) models for edge devices.
SOTA Medical and Healthcare text-only Small Language Models for Edge deployment
A Large-Scale, High-Quality Open Synthetic STEM Corpus for Efficient Language Model Pre-Training
-
QVAC Genesis III: A Large-Scale, High-Quality Open Synthetic STEM Corpus for Efficient Language Model Pre-Training
Paper • 2609.19513 • Published -
qvac/GenesisIII
Viewer • Updated • 319M • 2.06k • 1 -
qvac/qvac-genesis-iii-qwen3-1.7b-combined
Text Generation • 2B • Updated • 30 -
qvac/qvac-genesis-iii-qwen3-1.7b-fa
Text Generation • 2B • Updated • 13
SOTA General-Purpose Multimodal Vision-Language Models for Edge deployment
-
qvac/VisionPsy-Nano-460M
Image-Text-to-Text • 0.5B • Updated • 387 • 23 -
qvac/VisionPsy-Nano-460M-Flash
Image-Text-to-Text • 0.5B • Updated • 1.15k • 9 -
qvac/VisionPsy-Nano-460M-Flash-GGUFs
Image-Text-to-Text • 0.4B • Updated • 7.64k • 14 -
qvac/VisionPsy-Nano-460M-GGUFs
Image-Text-to-Text • 0.4B • Updated • 5.49k • 5
SOTA Small Language Models (SLMs) for fast and efficient African Machine Translation.
-
TranslatePsy-AfriSLM: High-Quality Data Scaling For Low-Resource Machine Translation
Paper • 2608.18655 • Published • 5 -
qvac/TranslatePsy-AfriSLM-Synthetic-Mix
Viewer • Updated • 206M • 1.56k • 4 -
qvac/TranslatePsy-AfriSLM-0.8B
Text Generation • 1B • Updated • 415 • 1 -
qvac/TranslatePsy-AfriSLM-0.8B-Q8-GGUF
Text Generation • 1B • Updated • 129 • 1
Compact multilingual Automatic Speech Recognition (ASR) model that transcribes speech to text, and purpose-built to run on the phone in your pocket.
-
qvac/TranscribePsy-250M-EU-sub-gguf
Automatic Speech Recognition • 0.3B • Updated • 232 • 3 -
qvac/TranscribePsy-250M-EU
Automatic Speech Recognition • 0.3B • Updated • 73 • 4 -
qvac/TranscribePsy-250M-EU-sub-coreml
Automatic Speech Recognition • 14.6M • Updated • 44 • 2 -
qvac/TranscribePsy-250M-EU-sub
Automatic Speech Recognition • 0.3B • Updated • 34 • 2
A Large-Scale, High-Quality Open Synthetic STEM Corpus for Efficient Language Model Pre-Training
-
QVAC Genesis III: A Large-Scale, High-Quality Open Synthetic STEM Corpus for Efficient Language Model Pre-Training
Paper • 2609.19513 • Published -
qvac/GenesisIII
Viewer • Updated • 319M • 2.06k • 1 -
qvac/qvac-genesis-iii-qwen3-1.7b-combined
Text Generation • 2B • Updated • 30 -
qvac/qvac-genesis-iii-qwen3-1.7b-fa
Text Generation • 2B • Updated • 13
Compact multilingual sequence to sequence neural machine translation (NMT) models for edge devices.
SOTA General-Purpose Multimodal Vision-Language Models for Edge deployment
-
qvac/VisionPsy-Nano-460M
Image-Text-to-Text • 0.5B • Updated • 387 • 23 -
qvac/VisionPsy-Nano-460M-Flash
Image-Text-to-Text • 0.5B • Updated • 1.15k • 9 -
qvac/VisionPsy-Nano-460M-Flash-GGUFs
Image-Text-to-Text • 0.4B • Updated • 7.64k • 14 -
qvac/VisionPsy-Nano-460M-GGUFs
Image-Text-to-Text • 0.4B • Updated • 5.49k • 5
SOTA Medical and Healthcare text-only Small Language Models for Edge deployment
SOTA Small Language Models (SLMs) for fast and efficient African Machine Translation.
-
TranslatePsy-AfriSLM: High-Quality Data Scaling For Low-Resource Machine Translation
Paper • 2608.18655 • Published • 5 -
qvac/TranslatePsy-AfriSLM-Synthetic-Mix
Viewer • Updated • 206M • 1.56k • 4 -
qvac/TranslatePsy-AfriSLM-0.8B
Text Generation • 1B • Updated • 415 • 1 -
qvac/TranslatePsy-AfriSLM-0.8B-Q8-GGUF
Text Generation • 1B • Updated • 129 • 1