[GH-ISSUE #531] What is the difference between the v1 and v2 models? #303

New issue

Closed

opened 2026-10-20 09:57:50 +00:00 by bsakaguchi · 5 comments

bsakaguchi commented

2026-10-20 09:57:50 +00:00

Owner

Originally created by @loboere on GitHub (Jun 14, 2023).
Original GitHub issue: https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/issues/531

Should I retrain the models to v2 for better quality?

Originally created by @loboere on GitHub (Jun 14, 2023). Original GitHub issue: https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/issues/531 Should I retrain the models to v2 for better quality?

bsakaguchi

2026-10-20 09:57:50 +00:00

closed this issue
added the
stale
label

bsakaguchi commented

2026-10-20 09:57:50 +00:00

Author

Owner

@RVC-Boss commented on GitHub (Jun 15, 2023):

Theoretically better quality.
You can have a try, it doesn't cost much.

@RVC-Boss commented on GitHub (Jun 15, 2023): Theoretically better quality. You can have a try, it doesn't cost much.

bsakaguchi commented

2026-10-20 09:57:51 +00:00

Author

Owner

@loboere commented on GitHub (Jun 15, 2023):

mm ok, is it possible to convert models v1 to v2 without training from 0?mm ok, is it possible to convert models v1 to v2 without training from 0?

@loboere commented on GitHub (Jun 15, 2023): mm ok, is it possible to convert models v1 to v2 without training from 0?mm ok, is it possible to convert models v1 to v2 without training from 0?

bsakaguchi commented

2026-10-20 09:57:51 +00:00

Author

Owner

@RVC-Boss commented on GitHub (Jun 15, 2023):

No, the input feature is not the same.

@RVC-Boss commented on GitHub (Jun 15, 2023): No, the input feature is not the same.

bsakaguchi commented

2026-10-20 09:57:51 +00:00

Author

Owner

@niizam commented on GitHub (Jun 17, 2023):

@RVC-Boss I would like to ask about training the RVC V2 model, which requires more VRAM compared to training the RVC V1 model. With a 4GB VRAM, the RVC V1 model is sufficient for training with a batch size of 1 on the GPU. However, it's not the case with the RVC V2 model. Is it possible to optimize it?

@niizam commented on GitHub (Jun 17, 2023): @RVC-Boss I would like to ask about training the RVC V2 model, which requires more VRAM compared to training the RVC V1 model. With a 4GB VRAM, the RVC V1 model is sufficient for training with a batch size of 1 on the GPU. However, it's not the case with the RVC V2 model. Is it possible to optimize it?

bsakaguchi commented

2026-10-20 09:57:51 +00:00

Author

Owner

@github-actions[bot] commented on GitHub (Apr 28, 2024):

This issue was closed because it has been inactive for 15 days since being marked as stale.

@github-actions[bot] commented on GitHub (Apr 28, 2024): This issue was closed because it has been inactive for 15 days since being marked as stale.

bsakaguchi referenced this issue

2026-10-20 10:14:13 +00:00

[GH-ISSUE #2860] error #2060