[Model] Add Ming-omni-tts MoE 16.8B-A3B + CFM CUDAGraph (#4341)

Signed-off-by: LHXuuu <xulianhao.xlh@antgroup.com>
Signed-off-by: Yuanheng Zhao <jonathan.zhaoyh@gmail.com>
Signed-off-by: Hongsheng Liu <liuhongsheng4@huawei.com>
Co-authored-by: Yuanheng Zhao <jonathan.zhaoyh@gmail.com>
Co-authored-by: Yuanheng Zhao <54058983+yuanheng-zhao@users.noreply.github.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
This commit is contained in:
authored and GitHub committed 2026-06-25 22:11:35 +08:00
1 parent 5db89b41c4
commit 0b04f2a0d6
17 files changed
+637 -57

No files matched your search

@@ -346,11 +346,8 @@ class AudioVAE(PreTrainedModel):
h, _ = self.encoder(waveform)
h = h.transpose(1, 2)
mean, logvar = torch.chunk(h, 2, dim=1)
logvar = torch.clamp(logvar, -30.0, 20.0)
std = torch.exp(0.5 * logvar)
latent = mean + std * torch.randn_like(mean)
latent = latent.transpose(1, 2)
mean = torch.chunk(h, 2, dim=1)[0]
latent = mean.transpose(1, 2)
return latent, frame_num
@torch.inference_mode()