From b5e5bc2ed73ab4b30c13c0e5b33f941e1e8f23f4 Mon Sep 17 00:00:00 2001
From: octo-patch <266937838+octo-patch@users.noreply.github.com>
Date: Fri, 17 Jul 2026 10:48:56 +0800
Subject: [PATCH] Add MiniMax music generation provider

---
 sdk/python/agentfield/media_providers.py | 42 ++++++++++++++++++++++++
 1 file changed, 42 insertions(+)

diff --git a/sdk/python/agentfield/media_providers.py b/sdk/python/agentfield/media_providers.py
index 2bbec0607..10123c038 100644
--- a/sdk/python/agentfield/media_providers.py
+++ b/sdk/python/agentfield/media_providers.py
@@ -152,6 +152,47 @@ async def generate_music(
         raise NotImplementedError(f"{self.name} does not support music generation")
 
 
+class MiniMaxProvider(MediaProvider):
+    """MiniMax music generation provider for global and China endpoints."""
+
+    def __init__(self, api_key: Optional[str] = None, region: str = "global_en"):
+        self._api_key = api_key
+        self.region = region
+
+    @property
+    def name(self) -> str:
+        return "minimax"
+
+    @property
+    def supported_modalities(self) -> List[str]:
+        return ["music"]
+
+    async def generate_image(self, prompt: str, model: Optional[str] = None, size: str = "1024x1024", quality: str = "standard", **kwargs) -> MultimodalResponse:
+        raise NotImplementedError(f"{self.name} does not support image generation")
+
+    async def generate_audio(self, text: str, model: Optional[str] = None, voice: str = "alloy", format: str = "wav", *, system: Optional[str] = None, **kwargs) -> MultimodalResponse:
+        raise NotImplementedError(f"{self.name} does not support speech generation")
+
+    async def generate_music(self, prompt: str, model: Optional[str] = None, duration: Optional[int] = None, **kwargs) -> MultimodalResponse:
+        import base64, os, aiohttp
+        key = self._api_key or os.environ.get("MINIMAX_API_KEY", "")
+        if not key:
+            raise ValueError("MiniMax API key required. Set MINIMAX_API_KEY or pass api_key.")
+        endpoint = "https://api.minimaxi.com/v1/music_generation" if self.region == "cn_zh" else "https://api.minimax.io/v1/music_generation"
+        body = {"model": model or "music-3.0", "prompt": prompt, "output_format": kwargs.pop("output_format", "url"), "stream": kwargs.pop("stream", False)}
+        if duration is not None:
+            body["audio_setting"] = {"sample_rate": kwargs.pop("sample_rate", 44100), "bitrate": kwargs.pop("bitrate", 256000), "format": kwargs.pop("format", "mp3"), "duration": duration}
+        body.update({k: v for k, v in kwargs.items() if k in {"lyrics", "lyrics_optimizer", "is_instrumental", "audio_setting", "aigc_watermark", "cover_feature_id"}})
+        async with aiohttp.ClientSession() as session:
+            async with session.post(endpoint, json=body, headers={"Authorization": f"Bearer {key}", "Content-Type": "application/json"}) as response:
+                data = await response.json()
+                if response.status >= 400 or data.get("base_resp", {}).get("status_code") != 0:
+                    raise RuntimeError(f"MiniMax music generation failed ({response.status}): {data}")
+        audio = data.get("data", {}).get("audio")
+        output = AudioOutput(data=audio if audio and body["output_format"] == "hex" else None, format=body.get("audio_setting", {}).get("format", "mp3"), url=audio if body["output_format"] == "url" else None)
+        return MultimodalResponse(text=prompt, audio=output, images=[], files=[], raw_response=data)
+
+
 class FalProvider(MediaProvider):
     """
     Fal.ai provider for image, audio, and video generation.
@@ -1654,6 +1695,7 @@ async def generate_music(
     "fal": FalProvider,
     "litellm": LiteLLMProvider,
     "openrouter": OpenRouterProvider,
+    "minimax": MiniMaxProvider,
 }