diff --git a/functions/tools/muapi_image_video_generator.py b/functions/tools/muapi_image_video_generator.py
new file mode 100644
index 0000000..711c9ea
--- /dev/null
+++ b/functions/tools/muapi_image_video_generator.py
@@ -0,0 +1,144 @@
+"""
+title: MuAPI Image & Video Generator
+author: muapi
+author_url: https://muapi.ai
+funding_url: https://muapi.ai
+version: 1.0.0
+license: MIT
+requirements: requests
+"""
+
+import asyncio
+import time
+import requests
+from typing import Optional
+from pydantic import BaseModel, Field
+
+BASE_URL = "https://api.muapi.ai/api/v1"
+
+IMAGE_MODELS = [
+    "flux-schnell", "flux-dev", "flux-kontext-pro", "hidream-fast",
+    "midjourney", "gpt4o", "imagen4", "imagen4-fast", "seedream",
+    "reve", "ideogram", "gpt-image-2",
+]
+
+VIDEO_MODELS = [
+    "veo3-fast", "veo3", "kling-master", "wan2.1", "wan2.2",
+    "seedance-pro", "runway", "pixverse", "sora",
+]
+
+
+class Tools:
+    class Valves(BaseModel):
+        MUAPI_API_KEY: str = Field(
+            default="",
+            description="Your MuAPI API key. Get one at https://muapi.ai/dashboard/api-keys",
+            json_schema_extra={"ui": {"label": "MuAPI API Key", "type": "password"}},
+        )
+        DEFAULT_IMAGE_MODEL: str = Field(
+            default="flux-schnell",
+            description=f"Default image model. Options: {', '.join(IMAGE_MODELS)}",
+        )
+        DEFAULT_VIDEO_MODEL: str = Field(
+            default="veo3-fast",
+            description=f"Default video model. Options: {', '.join(VIDEO_MODELS)}",
+        )
+        POLL_INTERVAL: float = Field(default=3.0, description="Seconds between status polls")
+        MAX_WAIT: int = Field(default=300, description="Max seconds to wait for generation")
+
+    def __init__(self):
+        self.valves = self.Valves()
+
+    def _submit_and_poll(self, endpoint: str, payload: dict) -> str:
+        """Submit a muapi.ai job and poll until completion."""
+        headers = {"x-api-key": self.valves.MUAPI_API_KEY, "Content-Type": "application/json"}
+
+        resp = requests.post(f"{BASE_URL}/{endpoint}", json=payload, headers=headers, timeout=30)
+        resp.raise_for_status()
+        request_id = resp.json()["request_id"]
+
+        deadline = time.time() + self.valves.MAX_WAIT
+        while time.time() < deadline:
+            time.sleep(self.valves.POLL_INTERVAL)
+            poll = requests.get(
+                f"{BASE_URL}/predictions/{request_id}/result",
+                headers={"x-api-key": self.valves.MUAPI_API_KEY},
+                timeout=15,
+            )
+            poll.raise_for_status()
+            data = poll.json()
+            if data["status"] == "completed":
+                outputs = data.get("outputs", [])
+                if not outputs:
+                    raise RuntimeError("Generation completed but returned no outputs")
+                return outputs[0]
+            if data["status"] in ("failed", "cancelled"):
+                raise RuntimeError(f"Generation {data['status']}: {data.get('error', '')}")
+
+        raise TimeoutError(f"Generation timed out after {self.valves.MAX_WAIT}s")
+
+    def generate_image(
+        self,
+        prompt: str,
+        model: Optional[str] = None,
+        width: Optional[int] = None,
+        height: Optional[int] = None,
+    ) -> str:
+        """
+        Generate an image from a text description using MuAPI.
+        Supports Flux, Midjourney, GPT-4o Image, Google Imagen 4, Seedream, and more.
+        Returns a markdown image with the result URL.
+
+        :param prompt: Text description of the image to generate
+        :param model: Model ID (e.g. flux-schnell, midjourney, gpt4o, imagen4). Uses default if not set.
+        :param width: Image width in pixels (optional)
+        :param height: Image height in pixels (optional)
+        """
+        if not self.valves.MUAPI_API_KEY:
+            return "Error: MUAPI_API_KEY is not set. Go to Admin > Tools and add your key."
+
+        endpoint = model or self.valves.DEFAULT_IMAGE_MODEL
+        payload: dict = {"prompt": prompt}
+        if width:
+            payload["width"] = width
+        if height:
+            payload["height"] = height
+
+        try:
+            url = self._submit_and_poll(endpoint, payload)
+            return f"![Generated image]({url})\n\n**Model:** {endpoint} | **Prompt:** {prompt}"
+        except Exception as e:
+            return f"Image generation failed: {e}"
+
+    def generate_video(
+        self,
+        prompt: str,
+        model: Optional[str] = None,
+        duration: int = 5,
+        aspect_ratio: str = "16:9",
+    ) -> str:
+        """
+        Generate a short video from a text description using MuAPI.
+        Supports Veo3, Kling, Wan, Seedance, Runway, Pixverse, Sora, and more.
+        Returns the video URL.
+
+        :param prompt: Text description of the video to generate
+        :param model: Model ID (e.g. veo3-fast, kling-master, wan2.1, runway). Uses default if not set.
+        :param duration: Duration in seconds (3-60, default 5)
+        :param aspect_ratio: Aspect ratio (16:9, 9:16, 1:1), default 16:9
+        """
+        if not self.valves.MUAPI_API_KEY:
+            return "Error: MUAPI_API_KEY is not set. Go to Admin > Tools and add your key."
+
+        endpoint = model or self.valves.DEFAULT_VIDEO_MODEL
+        payload: dict = {
+            "prompt": prompt,
+            "duration": duration,
+            "aspect_ratio": aspect_ratio,
+        }
+
+        try:
+            url = self._submit_and_poll(endpoint, payload)
+            return f"**Generated video:** {url}\n\n**Model:** {endpoint} | **Prompt:** {prompt}"
+        except Exception as e:
+            return f"Video generation failed: {e}"