Merge 76719afe7c into 985fb9d6ad

comfy/sd.py

@@ -469,9 +469,6 @@ class CLIP:
     def decode(self, token_ids, skip_special_tokens=True):
         return self.tokenizer.decode(token_ids, skip_special_tokens=skip_special_tokens)
 
-    def is_dynamic(self):
-        return self.patcher.is_dynamic()
-
 class VAE:
     def __init__(self, sd=None, device=None, config=None, dtype=None, metadata=None):
         is_seedvr2_vae = "decoder.up_blocks.2.upsamplers.0.upscale_conv.weight" in sd
@@ -1318,8 +1315,6 @@ class VAE:
         except:
             return None
 
-    def is_dynamic(self):
-        return self.patcher.is_dynamic()
 
 class StyleModel:
     def __init__(self, model, device="cpu"):

comfy/text_encoders/llama.py

@@ -937,41 +937,22 @@ class BaseGenerate:
             return torch.argmax(logits, dim=-1, keepdim=True)
 
         # Sampling mode
-        if len(token_history) > 0 and (repetition_penalty != 1.0 or (presence_penalty is not None and presence_penalty != 0.0)):
-            token_ids = torch.tensor(list(set(token_history)), device=logits.device)
-            token_logits = logits[:, token_ids]
-            if repetition_penalty != 1.0:
-                token_logits = torch.where(token_logits < 0, token_logits * repetition_penalty, token_logits / repetition_penalty)
-            if presence_penalty is not None and presence_penalty != 0.0:
-                token_logits = token_logits - presence_penalty
-            logits[:, token_ids] = token_logits
+        if repetition_penalty != 1.0:
+            for i in range(logits.shape[0]):
+                for token_id in set(token_history):
+                    logits[i, token_id] *= repetition_penalty if logits[i, token_id] < 0 else 1/repetition_penalty
+
+        if presence_penalty is not None and presence_penalty != 0.0:
+            for i in range(logits.shape[0]):
+                for token_id in set(token_history):
+                    logits[i, token_id] -= presence_penalty
 
         if temperature != 1.0:
             logits = logits / temperature
 
         if top_k > 0:
-            top_k = min(top_k, logits.shape[-1])
-            logits, top_indices = torch.topk(logits, top_k)
-
-            if min_p > 0.0:
-                probs_before_filter = torch.nn.functional.softmax(logits, dim=-1)
-                top_probs, _ = probs_before_filter.max(dim=-1, keepdim=True)
-                min_threshold = min_p * top_probs
-                indices_to_remove = probs_before_filter < min_threshold
-                logits[indices_to_remove] = torch.finfo(logits.dtype).min
-
-            if top_p < 1.0:
-                sorted_logits, sorted_indices = torch.sort(logits, descending=True)
-                cumulative_probs = torch.cumsum(torch.nn.functional.softmax(sorted_logits, dim=-1), dim=-1)
-                sorted_indices_to_remove = cumulative_probs > top_p
-                sorted_indices_to_remove[..., 0] = False
-                indices_to_remove = torch.zeros_like(logits, dtype=torch.bool)
-                indices_to_remove.scatter_(1, sorted_indices, sorted_indices_to_remove)
-                logits[indices_to_remove] = torch.finfo(logits.dtype).min
-
-            probs = torch.nn.functional.softmax(logits, dim=-1)
-            next_token = torch.multinomial(probs, num_samples=1, generator=generator)
-            return top_indices.gather(1, next_token)
+            indices_to_remove = logits < torch.topk(logits, top_k)[0][..., -1, None]
+            logits[indices_to_remove] = torch.finfo(logits.dtype).min
 
         if min_p > 0.0:
             probs_before_filter = torch.nn.functional.softmax(logits, dim=-1)

comfy_execution/caching.py

@@ -503,21 +503,6 @@ RAM_CACHE_DEFAULT_RAM_USAGE = 0.05
 
 RAM_CACHE_OLD_WORKFLOW_OOM_MULTIPLIER = 1.3
 
-
-def all_outputs_dynamic(outputs):
-    if outputs is None:
-        return False
-
-    for output in outputs:
-        if isinstance(output, (list, tuple)):
-            if not all_outputs_dynamic(output):
-                return False
-        elif not hasattr(output, "is_dynamic") or not output.is_dynamic():
-            return False
-
-    return True
-
-
 class RAMPressureCache(LRUCache):
 
     def __init__(self, key_class, enable_providers=False):
@@ -548,11 +533,7 @@ class RAMPressureCache(LRUCache):
         for key, cache_entry in self.cache.items():
             if not free_active and self.used_generation[key] == self.generation:
                 continue
-
-            if all_outputs_dynamic(cache_entry.outputs) and self.used_generation[key] == self.generation:
-                continue
-
-            oom_score = RAM_CACHE_OLD_WORKFLOW_OOM_MULTIPLIER ** (self.generation - self.used_generation[key])
+            oom_score =  RAM_CACHE_OLD_WORKFLOW_OOM_MULTIPLIER ** (self.generation - self.used_generation[key])
 
             ram_usage = RAM_CACHE_DEFAULT_RAM_USAGE
             def scan_list_for_ram_usage(outputs):

requirements.txt

@@ -1,6 +1,6 @@
 comfyui-frontend-package==1.45.20
 comfyui-workflow-templates==0.11.2
-comfyui-embedded-docs==0.5.7
+comfyui-embedded-docs==0.5.6
 torch
 torchsde
 torchvision