update examples

examples/flux2/model_inference_low_vram/FLUX.2-klein-4B.py

@@ -21,6 +21,7 @@ pipe = Flux2ImagePipeline.from_pretrained(
         ModelConfig(model_id="black-forest-labs/FLUX.2-klein-4B", origin_file_pattern="vae/diffusion_pytorch_model.safetensors"),
     ],
     tokenizer_config=ModelConfig(model_id="black-forest-labs/FLUX.2-klein-4B", origin_file_pattern="tokenizer/"),
+    vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
 )
 prompt = "Masterpiece, best quality. Anime-style portrait of a woman in a blue dress, underwater, surrounded by colorful bubbles."
 image = pipe(prompt, seed=0, rand_device="cuda", num_inference_steps=4)

examples/flux2/model_inference_low_vram/FLUX.2-klein-9B.py

@@ -21,6 +21,7 @@ pipe = Flux2ImagePipeline.from_pretrained(
         ModelConfig(model_id="black-forest-labs/FLUX.2-klein-4B", origin_file_pattern="vae/diffusion_pytorch_model.safetensors"),
     ],
     tokenizer_config=ModelConfig(model_id="black-forest-labs/FLUX.2-klein-4B", origin_file_pattern="tokenizer/"),
+    vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
 )
 prompt = "Masterpiece, best quality. Anime-style portrait of a woman in a blue dress, underwater, surrounded by colorful bubbles."
 image = pipe(prompt, seed=0, rand_device="cuda", num_inference_steps=4)

examples/flux2/model_inference_low_vram/FLUX.2-klein-base-4B.py

@@ -21,6 +21,7 @@ pipe = Flux2ImagePipeline.from_pretrained(
         ModelConfig(model_id="black-forest-labs/FLUX.2-klein-4B", origin_file_pattern="vae/diffusion_pytorch_model.safetensors"),
     ],
     tokenizer_config=ModelConfig(model_id="black-forest-labs/FLUX.2-klein-4B", origin_file_pattern="tokenizer/"),
+    vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
 )
 prompt = "Masterpiece, best quality. Anime-style portrait of a woman in a blue dress, underwater, surrounded by colorful bubbles."
 image = pipe(prompt, seed=0, rand_device="cuda", num_inference_steps=50, cfg_scale=4)

examples/flux2/model_inference_low_vram/FLUX.2-klein-base-9B.py

@@ -21,6 +21,7 @@ pipe = Flux2ImagePipeline.from_pretrained(
         ModelConfig(model_id="black-forest-labs/FLUX.2-klein-9B", origin_file_pattern="vae/diffusion_pytorch_model.safetensors"),
     ],
     tokenizer_config=ModelConfig(model_id="black-forest-labs/FLUX.2-klein-9B", origin_file_pattern="tokenizer/"),
+    vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
 )
 prompt = "Masterpiece, best quality. Anime-style portrait of a woman in a blue dress, underwater, surrounded by colorful bubbles."
 image = pipe(prompt, seed=0, rand_device="cuda", num_inference_steps=50, cfg_scale=4)

examples/z_image/model_inference_low_vram/Z-Image-Omni-Base-i2L.py

@@ -33,6 +33,7 @@ pipe = ZImagePipeline.from_pretrained(
         ModelConfig(model_id="DiffSynth-Studio/Z-Image-Omni-Base-i2L", origin_file_pattern="model.safetensors", **vram_config),
     ],
     tokenizer_config=ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="tokenizer/"),
+    vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
 )
 
 # Load images

examples/z_image/model_inference_low_vram/Z-Image-Omni-Base.py

@@ -22,6 +22,7 @@ pipe = ZImagePipeline.from_pretrained(
         ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
     ],
     tokenizer_config=ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="tokenizer/"),
+    vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
 )
 prompt = "Young Chinese woman in red Hanfu, intricate embroidery. Impeccable makeup, red floral forehead pattern. Elaborate high bun, golden phoenix headdress, red flowers, beads. Holds round folding fan with lady, trees, bird. Neon lightning-bolt lamp (⚡️), bright yellow glow, above extended left palm. Soft-lit outdoor night background, silhouetted tiered pagoda (西安大雁塔), blurred colorful distant lights."
 image = pipe(prompt=prompt, seed=0, num_inference_steps=40, cfg_scale=4)

examples/z_image/model_inference_low_vram/Z-Image-Turbo-Fun-Controlnet-Tile-2.1-8steps.py

@@ -24,6 +24,7 @@ pipe = ZImagePipeline.from_pretrained(
         ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
     ],
     tokenizer_config=ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="tokenizer/"),
+    vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
 )
 
 dataset_snapshot_download(

examples/z_image/model_inference_low_vram/Z-Image-Turbo-Fun-Controlnet-Union-2.1-8steps.py

@@ -24,6 +24,7 @@ pipe = ZImagePipeline.from_pretrained(
         ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
     ],
     tokenizer_config=ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="tokenizer/"),
+    vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
 )
 
 # Control

examples/z_image/model_inference_low_vram/Z-Image-Turbo-Fun-Controlnet-Union-2.1.py

@@ -24,6 +24,7 @@ pipe = ZImagePipeline.from_pretrained(
         ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
     ],
     tokenizer_config=ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="tokenizer/"),
+    vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
 )
 
 # Control

examples/z_image/model_inference_low_vram/Z-Image-i2L.py

@@ -32,6 +32,7 @@ pipe = ZImagePipeline.from_pretrained(
         ModelConfig(model_id="DiffSynth-Studio/Z-Image-i2L", origin_file_pattern="model.safetensors", **vram_config),
     ],
     tokenizer_config=ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="tokenizer/"),
+    vram_limit=0,
 )
 
 # Load images

examples/z_image/model_inference_low_vram/Z-Image.py

@@ -20,6 +20,7 @@ pipe = ZImagePipeline.from_pretrained(
         ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="vae/diffusion_pytorch_model.safetensors", **vram_config),
     ],
     tokenizer_config=ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="tokenizer/"),
+    vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
 )
 prompt = "Young Chinese woman in red Hanfu, intricate embroidery. Impeccable makeup, red floral forehead pattern. Elaborate high bun, golden phoenix headdress, red flowers, beads. Holds round folding fan with lady, trees, bird. Neon lightning-bolt lamp (⚡️), bright yellow glow, above extended left palm. Soft-lit outdoor night background, silhouetted tiered pagoda (西安大雁塔), blurred colorful distant lights."
 image = pipe(prompt=prompt, seed=42, rand_device="cuda", num_inference_steps=50, cfg_scale=4)