Update app.py

app.py

@@ -1,116 +1,191 @@
-import gradio as gr
-import requests
+# main/app.py
+# -------------------------------------------------------------------
+# FLUX.1-Schnell Space — fixed wiring + proper Space launch settings
+# -------------------------------------------------------------------
+import os
 import io
 import random
-import os
-import time
+import requests
 from PIL import Image
-from deep_translator import GoogleTranslator
-import json
-
-# Project by Nymbo
+import gradio as gr
+from deep_translator import GoogleTranslator  # used opportunistically
 
+# -----------------------------
+# Config / constants (simple)
+# -----------------------------
 API_URL = "https://api-inference.huggingface.co/models/black-forest-labs/FLUX.1-schnell"
-API_TOKEN = os.getenv("HF_READ_TOKEN")
-headers = {"Authorization": f"Bearer {API_TOKEN}"}
-timeout = 100
+# Try a few common env var names so you don't get bitten by naming
+API_TOKEN = (
+    os.getenv("HF_READ_TOKEN")
+    or os.getenv("HF_TOKEN")
+    or os.getenv("HUGGINGFACEHUB_API_TOKEN")
+)
+HEADERS = {"Authorization": f"Bearer {API_TOKEN}"} if API_TOKEN else {}
+TIMEOUT = 100  # seconds
+
+# -----------------------------------------------
+# Helper: translate to EN only if it looks Cyrillic
+# (keeps behavior without forcing translation)
+# -----------------------------------------------
+def maybe_translate_to_en(text: str) -> str:
+    # Count Cyrillic letters; if enough, treat as Russian and translate.
+    try:
+        cyr_count = sum("а" <= c <= "я" or "А" <= c <= "Я" for c in text)
+        if cyr_count >= max(3, len(text) // 10):
+            return GoogleTranslator(source="ru", target="en").translate(text)
+    except Exception:
+        # If translator fails, just return original prompt
+        pass
+    return text
 
-# Function to query the API and return the generated image
-def query(prompt, is_negative=False, steps=30, cfg_scale=7, sampler="DPM++ 2M Karras", seed=-1, strength=0.7, width=1024, height=1024):
-    if prompt == "" or prompt is None:
+# -------------------------------------------------------
+# Core: generate image
+# - Signature now matches the UI components by POSITION
+# - We send only supported params to the Inference API
+# -------------------------------------------------------
+def query(
+    prompt: str,             # main user prompt (Textbox)
+    negative_prompt: str,    # negative prompt (Textbox)
+    steps: int,              # sampling steps (Slider)
+    cfg_scale: float,        # guidance scale (Slider)
+    sampler: str,            # UI-only; ignored by FLUX text2img
+    seed: int,               # -1 => random, else fixed
+    strength: float,         # UI-only here (img2img only), ignored
+    width: int,              # output width
+    height: int,             # output height
+):
+    # Guard empty input
+    if not prompt:
         return None
 
+    # Give each generation a small key for readable logs
     key = random.randint(0, 999)
-    
-    API_TOKEN = random.choice([os.getenv("HF_READ_TOKEN")])
-    headers = {"Authorization": f"Bearer {API_TOKEN}"}
-    
-    # Translate the prompt from Russian to English if necessary
-    prompt = GoogleTranslator(source='ru', target='en').translate(prompt)
-    print(f'\033[1mGeneration {key} translation:\033[0m {prompt}')
-
-    # Add some extra flair to the prompt
-    prompt = f"{prompt} | ultra detail, ultra elaboration, ultra quality, perfect."
-    print(f'\033[1mGeneration {key}:\033[0m {prompt}')
-    
-    # Prepare the payload for the API call, including width and height
-    payload = {
-        "inputs": prompt,
-        "is_negative": is_negative,
-        "steps": steps,
-        "cfg_scale": cfg_scale,
-        "seed": seed if seed != -1 else random.randint(1, 1000000000),
-        "strength": strength,
-        "parameters": {
-            "width": width,  # Pass the width to the API
-            "height": height  # Pass the height to the API
-        }
+
+    # Optional translation + your style suffix
+    prompt_en = maybe_translate_to_en(prompt)
+    prompt_en = f"{prompt_en} | ultra detail, ultra elaboration, ultra quality, perfect."
+    print(f"\033[1mGeneration {key}:\033[0m {prompt_en}")
+
+    # Build payload using the names expected by the HF Inference API
+    params = {
+        "num_inference_steps": int(steps),
+        "guidance_scale": float(cfg_scale),
+        "width": int(width),
+        "height": int(height),
     }
+    if negative_prompt:
+        params["negative_prompt"] = negative_prompt
+    if int(seed) >= 0:
+        params["seed"] = int(seed)
+
+    # Make the request
+    resp = requests.post(
+        API_URL,
+        headers=HEADERS,
+        json={"inputs": prompt_en, "parameters": params},
+        timeout=TIMEOUT,
+    )
 
-    # Send the request to the API and handle the response
-    response = requests.post(API_URL, headers=headers, json=payload, timeout=timeout)
-    if response.status_code != 200:
-        print(f"Error: Failed to get image. Response status: {response.status_code}")
-        print(f"Response content: {response.text}")
-        if response.status_code == 503:
-            raise gr.Error(f"{response.status_code} : The model is being loaded")
-        raise gr.Error(f"{response.status_code}")
-    
+    # Handle common error modes: model warming, JSON error body, etc.
+    if resp.status_code != 200:
+        try:
+            err = resp.json()
+        except Exception:
+            err = {"error": resp.text}
+        msg = err.get("error") or err.get("message") or str(err)
+        if resp.status_code == 503:
+            # Friendly message when model is spinning up
+            raise gr.Error("503: The model is warming up. Please try again in a moment.")
+        raise gr.Error(f"{resp.status_code}: {msg}")
+
+    # Turn returned bytes into a PIL image
     try:
-        # Convert the response content into an image
-        image_bytes = response.content
-        image = Image.open(io.BytesIO(image_bytes))
-        print(f'\033[1mGeneration {key} completed!\033[0m ({prompt})')
+        image = Image.open(io.BytesIO(resp.content)).convert("RGB")
+        print(f"\033[1mGeneration {key} completed!\033[0m")
         return image
     except Exception as e:
-        print(f"Error when trying to open the image: {e}")
-        return None
+        print("Error decoding image:", e)
+        raise gr.Error("Failed to decode image from the model response.")
 
-# CSS to style the app
+# -----------------------
+# Minimal CSS for layout
+# -----------------------
 css = """
-#app-container {
-    max-width: 800px;
-    margin-left: auto;
-    margin-right: auto;
-}
+#app-container { max-width: 800px; margin-left: auto; margin-right: auto; }
 """
 
-# Build the Gradio UI with Blocks
-with gr.Blocks(theme='Nymbo/Nymbo_Theme', css=css) as app:
-    # Add a title to the app
+# -----------------------
+# UI layout (Blocks API)
+# -----------------------
+with gr.Blocks(theme="Nymbo/Nymbo_Theme", css=css) as app:
+    # Title banner (simple HTML)
     gr.HTML("<center><h1>FLUX.1-Schnell</h1></center>")
-    
-    # Container for all the UI elements
+
+    # Main column
     with gr.Column(elem_id="app-container"):
-        # Add a text input for the main prompt
+        # Prompt area
         with gr.Row():
             with gr.Column(elem_id="prompt-container"):
                 with gr.Row():
-                    text_prompt = gr.Textbox(label="Prompt", placeholder="Enter a prompt here", lines=2, elem_id="prompt-text-input")
-                
-                # Accordion for advanced settings
+                    # Prompt Textbox (user prompt)
+                    text_prompt = gr.Textbox(
+                        label="Prompt",
+                        placeholder="Enter a prompt here",
+                        lines=2,
+                        elem_id="prompt-text-input",
+                    )
+
+                # Advanced settings tucked into an Accordion
                 with gr.Row():
                     with gr.Accordion("Advanced Settings", open=False):
-                        negative_prompt = gr.Textbox(label="Negative Prompt", placeholder="What should not be in the image", value="(deformed, distorted, disfigured), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation, misspellings, typos", lines=3, elem_id="negative-prompt-text-input")
+                        # Negative Prompt Textbox (string, not a boolean!)
+                        negative_prompt = gr.Textbox(
+                            label="Negative Prompt",
+                            placeholder="What should not be in the image",
+                            value="(deformed, distorted, disfigured), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation, misspellings, typos",
+                            lines=3,
+                            elem_id="negative-prompt-text-input",
+                        )
+                        # Size sliders
                         with gr.Row():
                             width = gr.Slider(label="Width", value=1024, minimum=64, maximum=1216, step=32)
                             height = gr.Slider(label="Height", value=1024, minimum=64, maximum=1216, step=32)
+                        # Quality knobs
                         steps = gr.Slider(label="Sampling steps", value=4, minimum=1, maximum=100, step=1)
                         cfg = gr.Slider(label="CFG Scale", value=7, minimum=1, maximum=20, step=1)
-                        strength = gr.Slider(label="Strength", value=0.7, minimum=0, maximum=1, step=0.001)
-                        seed = gr.Slider(label="Seed", value=-1, minimum=-1, maximum=1000000000, step=1) # Setting the seed to -1 will make it random
-                        method = gr.Radio(label="Sampling method", value="DPM++ 2M Karras", choices=["DPM++ 2M Karras", "DPM++ SDE Karras", "Euler", "Euler a", "Heun", "DDIM"])
+                        # Shown for completeness; not used for text2img
+                        strength = gr.Slider(label="Strength (img2img only)", value=0.7, minimum=0, maximum=1, step=0.001)
+                        # -1 => random seed each run
+                        seed = gr.Slider(label="Seed (-1 = random)", value=-1, minimum=-1, maximum=1_000_000_000, step=1)
+                        # UI-only (Flux Inference API ignores this)
+                        sampler = gr.Radio(
+                            label="Sampling method (UI only)",
+                            value="DPM++ 2M Karras",
+                            choices=["DPM++ 2M Karras", "DPM++ SDE Karras", "Euler", "Euler a", "Heun", "DDIM"],
+                        )
 
-        # Add a button to trigger the image generation
+        # Run button
         with gr.Row():
-            text_button = gr.Button("Run", variant='primary', elem_id="gen-button")
-        
-        # Image output area to display the generated image
+            text_button = gr.Button("Run", variant="primary", elem_id="gen-button")
+
+        # Image output
         with gr.Row():
             image_output = gr.Image(type="pil", label="Image Output", elem_id="gallery")
-        
-        # Bind the button to the query function with the added width and height inputs
-        text_button.click(query, inputs=[text_prompt, negative_prompt, steps, cfg, method, seed, strength, width, height], outputs=image_output)
 
-# Launch the Gradio app
-app.launch(show_api=False, share=False)
+        # Wire the click to the function with matching POSITIONS
+        text_button.click(
+            fn=query,
+            inputs=[text_prompt, negative_prompt, steps, cfg, sampler, seed, strength, width, height],
+            outputs=image_output,
+        )
+
+# ------------------------------------------------------
+# Proper Space launch (no share tunnel, bind container)
+# ------------------------------------------------------
+if __name__ == "__main__":
+    app.launch(
+        server_name="0.0.0.0",                   # make server reachable in the Space
+        server_port=int(os.getenv("PORT", "7860")),  # Hugging Face sets PORT
+        show_api=False,                          # we don't need /docs
+        share=False,                             # Spaces already reverse-proxy
+    )