Model reference · open weights

quipus-0.6-speechv2

quipus-0.6-speechv2 is an open-weight audio or speech model from hyperneuronAILabs. quipus-0.6-speechv2 (BF16) weighs 1.2 GB; the smallest configuration that runs it is RTX 3060 12 GB.

• Quipus-0.6-speechv2 is an open-source text-to-speech model developed by HyperneuronAI that generates natural speech from text.
• It supports all 22 officially recognized languages of India and is built on a Qwen3 backbone with 608 million parameters.
• The model is released under the MIT license and is intended for research, experimentation, and building voice AI applications.

Summary of the hyperneuronAILabs/quipus-0.6-speechv2 model card, 2026-10-01

Hugging Face ↗