File size: 5,838 Bytes
882e052
 
c8f1f54
4fe456a
882e052
 
087c578
c8f1f54
087c578
882e052
9aeab3c
882e052
 
 
 
9aeab3c
 
c8f1f54
087c578
882e052
e86d5f6
882e052
 
 
e86d5f6
 
 
 
 
 
 
 
460870c
 
 
 
 
882e052
9aeab3c
e86d5f6
 
 
 
 
087c578
e86d5f6
 
 
 
 
882e052
 
 
460870c
c8f1f54
9aeab3c
882e052
9aeab3c
882e052
087c578
 
 
 
 
 
 
 
 
9aeab3c
 
 
 
087c578
 
 
 
 
882e052
 
 
087c578
 
882e052
 
087c578
 
9aeab3c
087c578
 
882e052
 
087c578
882e052
9aeab3c
882e052
087c578
882e052
9aeab3c
087c578
 
e86d5f6
 
 
 
 
 
 
 
4fe456a
087c578
882e052
 
9aeab3c
882e052
 
 
087c578
882e052
 
087c578
 
 
 
882e052
087c578
 
 
 
 
 
9aeab3c
 
 
087c578
 
9aeab3c
087c578
 
 
 
9aeab3c
087c578
 
 
 
 
 
 
882e052
 
 
 
 
087c578
 
 
 
 
 
 
 
882e052
c8f1f54
 
9aeab3c
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
import os
import requests
import gradio as gr
from PIL import Image, ImageDraw, ImageFont
import io
import time
from concurrent.futures import ThreadPoolExecutor

# ===== CONFIGURATION =====
HF_API_TOKEN = os.environ.get("HF_API_TOKEN")
MODEL_NAME = "stabilityai/stable-diffusion-xl-base-1.0"  # Using SDXL
API_URL = f"https://api-inference.huggingface.co/models/{MODEL_NAME}"
headers = {"Authorization": f"Bearer {HF_API_TOKEN}"}
WATERMARK_TEXT = "SelamGPT"
MAX_RETRIES = 3
TIMEOUT = 60  # Increased for SDXL's longer processing
EXECUTOR = ThreadPoolExecutor(max_workers=2)

# ===== WATERMARK FUNCTION =====
def add_watermark(image_bytes):
    """Convert to PNG with medium quality before watermarking"""
    try:
        image = Image.open(io.BytesIO(image_bytes)).convert("RGB")
        
        # Save as medium-quality PNG to buffer
        png_buffer = io.BytesIO()
        image.save(png_buffer, format="PNG", optimize=True, quality=85)  # Medium quality
        png_buffer.seek(0)
        
        # Add watermark to the PNG
        watermarked_image = Image.open(png_buffer)
        draw = ImageDraw.Draw(watermarked_image)
        font_size = 24
        try:
            font = ImageFont.truetype("Roboto-Bold.ttf", font_size)
        except:
            font = ImageFont.load_default(font_size)
        
        text_width = draw.textlength(WATERMARK_TEXT, font=font)
        x = watermarked_image.width - text_width - 10
        y = watermarked_image.height - 34
        
        draw.text((x+1, y+1), WATERMARK_TEXT, font=font, fill=(0, 0, 0, 128))
        draw.text((x, y), WATERMARK_TEXT, font=font, fill=(255, 255, 255))
        
        # Return as PNG bytes
        final_buffer = io.BytesIO()
        watermarked_image.save(final_buffer, format="PNG", optimize=True, quality=85)
        final_buffer.seek(0)
        return Image.open(final_buffer)
        
    except Exception as e:
        print(f"Watermark error: {str(e)}")
        return Image.open(io.BytesIO(image_bytes))

# ===== IMAGE GENERATION (SDXL-OPTIMIZED) =====
def generate_image(prompt):
    """Generate image with SDXL-specific parameters"""
    if not prompt.strip():
        return None, "⚠️ Please enter a prompt"
    
    def api_call():
        return requests.post(
            API_URL,
            headers=headers,
            json={
                "inputs": prompt,
                "parameters": {
                    "height": 1024,  # SDXL's native resolution
                    "width": 1024,
                    "num_inference_steps": 30,  # Better quality than 25
                    "guidance_scale": 7.5  # SDXL's optimal value
                },
                "options": {"wait_for_model": True}
            },
            timeout=TIMEOUT
        )
    
    for attempt in range(MAX_RETRIES):
        try:
            future = EXECUTOR.submit(api_call)
            response = future.result()
            
            if response.status_code == 200:
                return add_watermark(response.content), "✔️ Generation successful"
            elif response.status_code == 503:
                wait_time = (attempt + 1) * 15  # Longer wait for SDXL
                print(f"Model loading, waiting {wait_time}s...")
                time.sleep(wait_time)
                continue
            else:
                return None, f"⚠️ API Error: {response.text[:200]}"
        except requests.Timeout:
            return None, f"⚠️ Timeout: Model took >{TIMEOUT}s to respond"
        except Exception as e:
            return None, f"⚠️ Unexpected error: {str(e)[:200]}"
    
    return None, "⚠️ Failed after multiple attempts. Please try later."

# ===== GRADIO INTERFACE =====
with gr.Blocks() as demo:
    
    output_image = gr.Image(
        label="Generated Image",
        type="pil",  # Force PIL/PNG output
        format="png",  # Explicit PNG format
        height=512
    )

with gr.Blocks(theme=theme, title="SelamGPT Image Generator") as demo:
    gr.Markdown("""
    # 🎨 SelamGPT Image Generator
    *Now powered by Stable Diffusion XL (1024x1024 resolution)*
    """)
    
    with gr.Row():
        with gr.Column(scale=3):
            prompt_input = gr.Textbox(
                label="Describe your image",
                placeholder="A futuristic Ethiopian city with flying cars...",
                lines=3,
                max_lines=5,
                elem_id="prompt-box"
            )
            with gr.Row():
                generate_btn = gr.Button("Generate Image", variant="primary")
                clear_btn = gr.Button("Clear")
            
            gr.Examples(
                examples=[
                    ["An ancient Aksumite warrior in cyberpunk armor, 4k detailed"],
                    ["Traditional Ethiopian coffee ceremony in zero gravity, photorealistic"],
                    ["Portrait of a Habesha queen with golden jewelry, studio lighting"]
                ],
                inputs=prompt_input,
                label="Try these SDXL-optimized prompts:"
            )
            
        with gr.Column(scale=2):
            output_image = gr.Image(
                label="Generated Image (1024x1024)",
                height=512,
                elem_id="output-image"
            )
            status_output = gr.Textbox(
                label="Status",
                interactive=False,
                elem_id="status-box"
            )
    
    generate_btn.click(
        fn=generate_image,
        inputs=prompt_input,
        outputs=[output_image, status_output],
        queue=True,
        show_progress="minimal"
    )
    
    clear_btn.click(
        fn=lambda: [None, ""],
        outputs=[output_image, status_output]
    )

if __name__ == "__main__":
    demo.queue(max_size=2)
    demo.launch(server_name="0.0.0.0", server_port=7860)