fixes to make the model work

comfy/ldm/hunyuan_foley/model.py

@@ -154,6 +154,7 @@ class ChannelLastConv1d(nn.Module):
         object.__setattr__(self, "_underlying", underlying)
     
     def forward(self, x: torch.Tensor) -> torch.Tensor:
+        self._underlying = self._underlying.to(x.dtype)
         x = self._underlying(x.permute(0, 2, 1))
         return x.permute(0, 2, 1)
 
@@ -219,6 +220,7 @@ class FinalLayer1D(nn.Module):
     def forward(self, x, c):
         shift, scale = self.adaLN_modulation(c).chunk(2, dim=-1)
         x = modulate(self.norm_final(x), shift=shift, scale=scale)
+        self.linear = self.linear.to(x.dtype)
         x = self.linear(x)
         return x
     
@@ -614,6 +616,7 @@ class SingleStreamBlock(nn.Module):
         modulation = self.modulation(cond)
         shift_msa, scale_msa, gate_msa, shift_mlp, scale_mlp, gate_mlp = modulation.chunk(6, dim=-1)
         x_norm1 = self.norm1(x) * (1 + scale_msa) + shift_msa
+        x_norm1 = x_norm1.to(next(self.linear_qkv.parameters()).dtype)
 
         qkv = self.linear_qkv(x_norm1)
         q, k, v = self.rearrange(qkv).chunk(3, dim=-1)
@@ -647,7 +650,11 @@ def find_period_by_first_row(mat):
     if not candidate_positions:
         return L
 
-    return len(mat[:candidate_positions[0]])
+    for p in sorted(candidate_positions):
+        a, b = mat[p:], mat[:-p]
+        if torch.equal(a, b):
+            return p
+    return L
 
 def trim_repeats(expanded):
     seq = expanded[0]
@@ -865,7 +872,14 @@ class HunyuanVideoFoley(nn.Module):
         
         uncond_1, uncond_2, cond_neg, clip_feat, sync_feat, cond_pos = [unlock_cpu_tensor(t, device) for t in (uncond_1, uncond_2, cond_neg, clip_feat, sync_feat, cond_pos)]
 
+        diff = cond_pos.shape[1] - cond_neg.shape[1]
+        if cond_neg.shape[1] < cond_pos.shape[1]:
+            cond_neg = torch.nn.functional.pad(cond_neg, (0, 0, 0, diff))
+        elif diff < 0:
+            cond_pos = torch.nn.functional.pad(cond_pos, (0, 0, 0, torch.abs(diff)))
+
         clip_feat, sync_feat, cond = torch.cat([uncond_1, clip_feat]), torch.cat([uncond_2, sync_feat]), torch.cat([cond_neg, cond_pos])
+        clip_feat = clip_feat.view(2, -1, 768)
 
         if drop_visual is not None:
             clip_feat[drop_visual] = self.get_empty_clip_sequence().to(dtype=clip_feat.dtype)
@@ -954,8 +968,7 @@ class HunyuanVideoFoley(nn.Module):
         audio = self.final_layer(audio, vec)
         audio = self.unpatchify1d(audio, tl)
 
-        uncond, cond = torch.chunk(2, audio)
-        return torch.cat([cond, uncond])
+        return audio
 
     def unpatchify1d(self, x, l):
         c = self.unpatchify_channels

comfy/ldm/hydit/posemb_layers.py

@@ -183,8 +183,10 @@ def get_1d_rotary_pos_embed(dim: int, pos: Union[np.ndarray, int], theta: float
 
     if freq_scaling != 1.0:
         freqs *= freq_scaling
-        
-    t = torch.from_numpy(pos).to(freqs.device)  # type: ignore  # [S]
+    if not isinstance(pos, torch.Tensor):
+        t = torch.from_numpy(pos).to(freqs.device)  # type: ignore  # [S]
+    else:
+        t = pos.to(freqs.device)
     freqs = torch.outer(t, freqs).float()  # type: ignore   # [S, D/2]
     if use_real:
         freqs_cos = freqs.cos().repeat_interleave(2, dim=1)  # [S, D]

comfy/text_encoders/clap_model.py

@@ -361,4 +361,4 @@ class ClapTextEncoderModel(sd1_clip.SDClipModel):
 class ClapLargeTokenizer(sd1_clip.SDTokenizer):
     def __init__(self, embedding_directory=None, tokenizer_data={}):
         tokenizer_path = os.path.join(os.path.dirname(os.path.realpath(__file__)), "clap_tokenizer")
-        super().__init__(tokenizer_path, pad_with_end=False, embedding_size=2048, embedding_key='clap_l', tokenizer_class=AutoTokenizer, has_start_token=False, has_end_token=False, pad_to_max_length=False, max_length=99999999, min_length=1, pad_token=1, tokenizer_data=tokenizer_data)
+        super().__init__(tokenizer_path, pad_with_end=False, embedding_size=2048, embedding_key='clap_l', tokenizer_class=AutoTokenizer, has_start_token=False, has_end_token=False, pad_to_max_length=False, max_length=77, min_length=1, pad_token=1, tokenizer_data=tokenizer_data)