update

app.py

@@ -44,11 +44,14 @@ def init():
     # The undistilled model that uses CFG ("pro") which can use negative prompts
     # was not released.
     bfl_repo = "cocktailpeanut/xulf-s"
+    te_repo = "comfyanonymous/flux_text_encoders"
 
     scheduler = FlowMatchEulerDiscreteScheduler.from_pretrained(bfl_repo, subfolder="scheduler")
-    text_encoder = CLIPTextModel.from_pretrained("openai/clip-vit-large-patch14", torch_dtype=dtype)
+    #text_encoder = CLIPTextModel.from_pretrained("openai/clip-vit-large-patch14", torch_dtype=dtype)
+    text_encoder = CLIPTextModel.from_pretrained("./flux_text_encoders/clip_l.safetensors", torch_dtype=dtype)
     tokenizer = CLIPTokenizer.from_pretrained("openai/clip-vit-large-patch14", torch_dtype=dtype)
-    text_encoder_2 = T5EncoderModel.from_pretrained(bfl_repo, subfolder="text_encoder_2", torch_dtype=dtype)
+    #text_encoder_2 = T5EncoderModel.from_pretrained(bfl_repo, subfolder="text_encoder_2", torch_dtype=dtype)
+    text_encoder_2 = T5EncoderModel.from_pretrained("./flux_text_encoders/t5xxl_fp8_e4m3fn.safetensors", torch_dtype=dtype)
     tokenizer_2 = T5TokenizerFast.from_pretrained(bfl_repo, subfolder="tokenizer_2", torch_dtype=dtype)
     vae = AutoencoderKL.from_pretrained(bfl_repo, subfolder="vae", torch_dtype=dtype)
     transformer = FluxTransformer2DModel.from_pretrained(bfl_repo, subfolder="transformer", torch_dtype=dtype)