Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -1,4 +1,5 @@
|
|
1 |
import torch
|
|
|
2 |
torch.backends.cuda.matmul.allow_tf32 = True
|
3 |
torch.backends.cudnn.allow_tf32 = True
|
4 |
import gradio as gr
|
@@ -25,12 +26,17 @@ pipe = FluxWithCFGPipeline.from_pretrained(
|
|
25 |
)
|
26 |
pipe.vae = AutoencoderTiny.from_pretrained("madebyollin/taef1", torch_dtype=dtype)
|
27 |
pipe.to("cuda")
|
28 |
-
pipe.load_lora_weights(
|
|
|
|
|
|
|
|
|
29 |
pipe.set_adapters(["better"], adapter_weights=[1.0])
|
30 |
pipe.fuse_lora(adapter_name=["better"], lora_scale=1.0)
|
31 |
pipe.unload_lora_weights()
|
32 |
|
33 |
-
|
|
|
34 |
pipe.vae.to(memory_format=torch.channels_last)
|
35 |
|
36 |
pipe.enable_xformers_memory_efficient_attention()
|
@@ -39,7 +45,15 @@ torch.cuda.empty_cache()
|
|
39 |
|
40 |
# Inference function
|
41 |
@spaces.GPU(duration=25)
|
42 |
-
def generate_image(
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
43 |
if randomize_seed:
|
44 |
seed = random.randint(0, MAX_SEED)
|
45 |
generator = torch.Generator().manual_seed(int(float(seed)))
|
@@ -47,9 +61,15 @@ def generate_image(prompt, seed=24, width=DEFAULT_WIDTH, height=DEFAULT_HEIGHT,
|
|
47 |
start_time = time.time()
|
48 |
|
49 |
# Initialize static inputs for CUDA graph
|
50 |
-
static_latents = torch.randn(
|
51 |
-
|
52 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
53 |
static_text_ids = torch.tensor([[[1, 2, 3]]], dtype=torch.int32, device="cuda")
|
54 |
static_latent_image_ids = torch.tensor([1], dtype=torch.int64, device="cuda")
|
55 |
static_timestep = torch.tensor([999], dtype=dtype, device="cuda")
|
@@ -86,11 +106,22 @@ def generate_image(prompt, seed=24, width=DEFAULT_WIDTH, height=DEFAULT_HEIGHT,
|
|
86 |
joint_attention_kwargs=pipe.joint_attention_kwargs,
|
87 |
return_dict=False,
|
88 |
)[0]
|
89 |
-
static_latents_out = pipe.scheduler.step(
|
90 |
-
|
|
|
|
|
|
|
|
|
91 |
|
92 |
# Graph-based generation function
|
93 |
-
def generate_with_graph(
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
94 |
static_latents.copy_(latents)
|
95 |
static_prompt_embeds.copy_(prompt_embeds)
|
96 |
static_pooled_prompt_embeds.copy_(pooled_prompt_embeds)
|
@@ -101,15 +132,15 @@ def generate_image(prompt, seed=24, width=DEFAULT_WIDTH, height=DEFAULT_HEIGHT,
|
|
101 |
return static_output
|
102 |
|
103 |
# Only generate the last image in the sequence
|
104 |
-
img = pipe.generate_images(
|
105 |
-
|
106 |
-
|
107 |
-
|
108 |
-
|
109 |
-
|
110 |
-
|
111 |
-
|
112 |
-
latency = f"Latency: {(time.time()-start_time):.2f} seconds"
|
113 |
return img, seed, latency
|
114 |
|
115 |
# Example prompts
|
@@ -127,12 +158,18 @@ examples = [
|
|
127 |
with gr.Blocks() as demo:
|
128 |
with gr.Column(elem_id="app-container"):
|
129 |
gr.Markdown("# 🎨 Realtime FLUX Image Generator")
|
130 |
-
gr.Markdown(
|
131 |
-
|
|
|
|
|
|
|
|
|
132 |
|
133 |
with gr.Row():
|
134 |
with gr.Column(scale=2.5):
|
135 |
-
result = gr.Image(
|
|
|
|
|
136 |
with gr.Column(scale=1):
|
137 |
prompt = gr.Text(
|
138 |
label="Prompt",
|
@@ -146,15 +183,39 @@ with gr.Blocks() as demo:
|
|
146 |
|
147 |
with gr.Column("Advanced Options"):
|
148 |
with gr.Row():
|
149 |
-
realtime = gr.Checkbox(
|
|
|
|
|
|
|
|
|
150 |
latency = gr.Text(label="Latency")
|
151 |
with gr.Row():
|
152 |
seed = gr.Number(label="Seed", value=42)
|
153 |
-
randomize_seed = gr.Checkbox(
|
|
|
|
|
154 |
with gr.Row():
|
155 |
-
width = gr.Slider(
|
156 |
-
|
157 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
158 |
|
159 |
with gr.Row():
|
160 |
gr.Markdown("### 🌟 Inspiration Gallery")
|
@@ -164,7 +225,7 @@ with gr.Blocks() as demo:
|
|
164 |
fn=generate_image,
|
165 |
inputs=[prompt],
|
166 |
outputs=[result, seed, latency],
|
167 |
-
cache_examples="lazy"
|
168 |
)
|
169 |
|
170 |
enhanceBtn.click(
|
@@ -173,7 +234,7 @@ with gr.Blocks() as demo:
|
|
173 |
outputs=[result, seed, latency],
|
174 |
show_progress="full",
|
175 |
queue=False,
|
176 |
-
concurrency_limit=None
|
177 |
)
|
178 |
|
179 |
generateBtn.click(
|
@@ -182,13 +243,13 @@ with gr.Blocks() as demo:
|
|
182 |
outputs=[result, seed, latency],
|
183 |
show_progress="full",
|
184 |
api_name="RealtimeFlux",
|
185 |
-
queue=False
|
186 |
)
|
187 |
|
188 |
def update_ui(realtime_enabled):
|
189 |
return {
|
190 |
prompt: gr.update(interactive=True),
|
191 |
-
generateBtn: gr.update(visible=not realtime_enabled)
|
192 |
}
|
193 |
|
194 |
realtime.change(
|
@@ -196,7 +257,7 @@ with gr.Blocks() as demo:
|
|
196 |
inputs=[realtime],
|
197 |
outputs=[prompt, generateBtn],
|
198 |
queue=False,
|
199 |
-
concurrency_limit=None
|
200 |
)
|
201 |
|
202 |
async def realtime_generation(*args):
|
@@ -211,18 +272,26 @@ with gr.Blocks() as demo:
|
|
211 |
outputs=[result, seed, latency],
|
212 |
show_progress="full",
|
213 |
queue=False,
|
214 |
-
concurrency_limit=None
|
215 |
)
|
216 |
|
217 |
for component in [prompt, width, height, num_inference_steps]:
|
218 |
component.input(
|
219 |
fn=realtime_generation,
|
220 |
-
inputs=[
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
221 |
outputs=[result, seed, latency],
|
222 |
show_progress="hidden",
|
223 |
trigger_mode="always_last",
|
224 |
queue=True,
|
225 |
-
concurrency_limit=None
|
226 |
)
|
227 |
|
228 |
# Launch the app
|
|
|
1 |
import torch
|
2 |
+
|
3 |
torch.backends.cuda.matmul.allow_tf32 = True
|
4 |
torch.backends.cudnn.allow_tf32 = True
|
5 |
import gradio as gr
|
|
|
26 |
)
|
27 |
pipe.vae = AutoencoderTiny.from_pretrained("madebyollin/taef1", torch_dtype=dtype)
|
28 |
pipe.to("cuda")
|
29 |
+
pipe.load_lora_weights(
|
30 |
+
"hugovntr/flux-schnell-realism",
|
31 |
+
weight_name="schnell-realism_v2.3.safetensors",
|
32 |
+
adapter_name="better",
|
33 |
+
)
|
34 |
pipe.set_adapters(["better"], adapter_weights=[1.0])
|
35 |
pipe.fuse_lora(adapter_name=["better"], lora_scale=1.0)
|
36 |
pipe.unload_lora_weights()
|
37 |
|
38 |
+
# Corrected: Access 'transformer' instead of 'unet'
|
39 |
+
pipe.transformer.to(memory_format=torch.channels_last)
|
40 |
pipe.vae.to(memory_format=torch.channels_last)
|
41 |
|
42 |
pipe.enable_xformers_memory_efficient_attention()
|
|
|
45 |
|
46 |
# Inference function
|
47 |
@spaces.GPU(duration=25)
|
48 |
+
def generate_image(
|
49 |
+
prompt,
|
50 |
+
seed=24,
|
51 |
+
width=DEFAULT_WIDTH,
|
52 |
+
height=DEFAULT_HEIGHT,
|
53 |
+
randomize_seed=False,
|
54 |
+
num_inference_steps=2,
|
55 |
+
progress=gr.Progress(track_tqdm=True),
|
56 |
+
):
|
57 |
if randomize_seed:
|
58 |
seed = random.randint(0, MAX_SEED)
|
59 |
generator = torch.Generator().manual_seed(int(float(seed)))
|
|
|
61 |
start_time = time.time()
|
62 |
|
63 |
# Initialize static inputs for CUDA graph
|
64 |
+
static_latents = torch.randn(
|
65 |
+
(1, 4, height // 8, width // 8), dtype=dtype, device="cuda"
|
66 |
+
)
|
67 |
+
static_prompt_embeds = torch.randn(
|
68 |
+
(2, 77, 768), dtype=dtype, device="cuda"
|
69 |
+
) # Adjust dimensions as needed
|
70 |
+
static_pooled_prompt_embeds = torch.randn(
|
71 |
+
(2, 768), dtype=dtype, device="cuda"
|
72 |
+
) # Adjust dimensions as needed
|
73 |
static_text_ids = torch.tensor([[[1, 2, 3]]], dtype=torch.int32, device="cuda")
|
74 |
static_latent_image_ids = torch.tensor([1], dtype=torch.int64, device="cuda")
|
75 |
static_timestep = torch.tensor([999], dtype=dtype, device="cuda")
|
|
|
106 |
joint_attention_kwargs=pipe.joint_attention_kwargs,
|
107 |
return_dict=False,
|
108 |
)[0]
|
109 |
+
static_latents_out = pipe.scheduler.step(
|
110 |
+
static_noise_pred, static_timestep, static_latents, return_dict=False
|
111 |
+
)[0]
|
112 |
+
static_output = pipe._decode_latents_to_image(
|
113 |
+
static_latents_out, height, width, "pil"
|
114 |
+
)
|
115 |
|
116 |
# Graph-based generation function
|
117 |
+
def generate_with_graph(
|
118 |
+
latents,
|
119 |
+
prompt_embeds,
|
120 |
+
pooled_prompt_embeds,
|
121 |
+
text_ids,
|
122 |
+
latent_image_ids,
|
123 |
+
timestep,
|
124 |
+
):
|
125 |
static_latents.copy_(latents)
|
126 |
static_prompt_embeds.copy_(prompt_embeds)
|
127 |
static_pooled_prompt_embeds.copy_(pooled_prompt_embeds)
|
|
|
132 |
return static_output
|
133 |
|
134 |
# Only generate the last image in the sequence
|
135 |
+
img = pipe.generate_images(
|
136 |
+
prompt=prompt,
|
137 |
+
width=width,
|
138 |
+
height=height,
|
139 |
+
num_inference_steps=num_inference_steps,
|
140 |
+
generator=generator,
|
141 |
+
generate_with_graph=generate_with_graph,
|
142 |
+
)
|
143 |
+
latency = f"Latency: {(time.time()-start_time):.2f} seconds"
|
144 |
return img, seed, latency
|
145 |
|
146 |
# Example prompts
|
|
|
158 |
with gr.Blocks() as demo:
|
159 |
with gr.Column(elem_id="app-container"):
|
160 |
gr.Markdown("# 🎨 Realtime FLUX Image Generator")
|
161 |
+
gr.Markdown(
|
162 |
+
"Generate stunning images in real-time with Modified Flux.Schnell pipeline."
|
163 |
+
)
|
164 |
+
gr.Markdown(
|
165 |
+
"<span style='color: red;'>Note: Sometimes it stucks or stops generating images (I don't know why). In that situation just refresh the site.</span>"
|
166 |
+
)
|
167 |
|
168 |
with gr.Row():
|
169 |
with gr.Column(scale=2.5):
|
170 |
+
result = gr.Image(
|
171 |
+
label="Generated Image", show_label=False, interactive=False
|
172 |
+
)
|
173 |
with gr.Column(scale=1):
|
174 |
prompt = gr.Text(
|
175 |
label="Prompt",
|
|
|
183 |
|
184 |
with gr.Column("Advanced Options"):
|
185 |
with gr.Row():
|
186 |
+
realtime = gr.Checkbox(
|
187 |
+
label="Realtime Toggler",
|
188 |
+
info="If TRUE then uses more GPU but create image in realtime.",
|
189 |
+
value=False,
|
190 |
+
)
|
191 |
latency = gr.Text(label="Latency")
|
192 |
with gr.Row():
|
193 |
seed = gr.Number(label="Seed", value=42)
|
194 |
+
randomize_seed = gr.Checkbox(
|
195 |
+
label="Randomize Seed", value=True
|
196 |
+
)
|
197 |
with gr.Row():
|
198 |
+
width = gr.Slider(
|
199 |
+
label="Width",
|
200 |
+
minimum=256,
|
201 |
+
maximum=MAX_IMAGE_SIZE,
|
202 |
+
step=32,
|
203 |
+
value=DEFAULT_WIDTH,
|
204 |
+
)
|
205 |
+
height = gr.Slider(
|
206 |
+
label="Height",
|
207 |
+
minimum=256,
|
208 |
+
maximum=MAX_IMAGE_SIZE,
|
209 |
+
step=32,
|
210 |
+
value=DEFAULT_HEIGHT,
|
211 |
+
)
|
212 |
+
num_inference_steps = gr.Slider(
|
213 |
+
label="Inference Steps",
|
214 |
+
minimum=1,
|
215 |
+
maximum=4,
|
216 |
+
step=1,
|
217 |
+
value=DEFAULT_INFERENCE_STEPS,
|
218 |
+
)
|
219 |
|
220 |
with gr.Row():
|
221 |
gr.Markdown("### 🌟 Inspiration Gallery")
|
|
|
225 |
fn=generate_image,
|
226 |
inputs=[prompt],
|
227 |
outputs=[result, seed, latency],
|
228 |
+
cache_examples="lazy",
|
229 |
)
|
230 |
|
231 |
enhanceBtn.click(
|
|
|
234 |
outputs=[result, seed, latency],
|
235 |
show_progress="full",
|
236 |
queue=False,
|
237 |
+
concurrency_limit=None,
|
238 |
)
|
239 |
|
240 |
generateBtn.click(
|
|
|
243 |
outputs=[result, seed, latency],
|
244 |
show_progress="full",
|
245 |
api_name="RealtimeFlux",
|
246 |
+
queue=False,
|
247 |
)
|
248 |
|
249 |
def update_ui(realtime_enabled):
|
250 |
return {
|
251 |
prompt: gr.update(interactive=True),
|
252 |
+
generateBtn: gr.update(visible=not realtime_enabled),
|
253 |
}
|
254 |
|
255 |
realtime.change(
|
|
|
257 |
inputs=[realtime],
|
258 |
outputs=[prompt, generateBtn],
|
259 |
queue=False,
|
260 |
+
concurrency_limit=None,
|
261 |
)
|
262 |
|
263 |
async def realtime_generation(*args):
|
|
|
272 |
outputs=[result, seed, latency],
|
273 |
show_progress="full",
|
274 |
queue=False,
|
275 |
+
concurrency_limit=None,
|
276 |
)
|
277 |
|
278 |
for component in [prompt, width, height, num_inference_steps]:
|
279 |
component.input(
|
280 |
fn=realtime_generation,
|
281 |
+
inputs=[
|
282 |
+
realtime,
|
283 |
+
prompt,
|
284 |
+
seed,
|
285 |
+
width,
|
286 |
+
height,
|
287 |
+
randomize_seed,
|
288 |
+
num_inference_steps,
|
289 |
+
],
|
290 |
outputs=[result, seed, latency],
|
291 |
show_progress="hidden",
|
292 |
trigger_mode="always_last",
|
293 |
queue=True,
|
294 |
+
concurrency_limit=None,
|
295 |
)
|
296 |
|
297 |
# Launch the app
|