[diffusion] http-server: relax openai image endpoint's strict content_type limit (#15717)

python/sglang/multimodal_gen/docs/openai_api.md

@@ -104,14 +104,15 @@ curl -sS -X POST "http://localhost:30010/v1/images/generations" \
 
 **Endpoint:** `POST /v1/images/edits`
 
-This endpoint accepts a multipart form upload with an input image and a text prompt. The server can return either a base64-encoded image or a URL to download the image.
+This endpoint accepts a multipart form upload with input images and a text prompt. The server can return either a base64-encoded image or a URL to download the image.
 
 **Curl Example (b64_json response):**
 
 ```bash
 curl -sS -X POST "http://localhost:30010/v1/images/edits" \
   -H "Authorization: Bearer sk-proj-1234567890" \
-  -F "image=@input.png" \
+  -F "image=@local_input_image.png" \
+  -F "url=image_url.jpg" \
   -F "prompt=A calico cat playing a piano on stage" \
   -F "size=1024x1024" \
   -F "response_format=b64_json"
@@ -122,7 +123,8 @@ curl -sS -X POST "http://localhost:30010/v1/images/edits" \
 ```bash
 curl -sS -X POST "http://localhost:30010/v1/images/edits" \
   -H "Authorization: Bearer sk-proj-1234567890" \
-  -F "image=@input.png" \
+  -F "image=@local_input_image.png" \
+  -F "url=image_url.jpg" \
   -F "prompt=A calico cat playing a piano on stage" \
   -F "size=1024x1024" \
   -F "response_format=url"

python/sglang/multimodal_gen/runtime/entrypoints/openai/utils.py

@@ -64,7 +64,7 @@ async def _save_upload_to_path(upload: UploadFile, target_path: str) -> str:
     return target_path
 
 
-async def _maybe_url_image(img_url: str, target_path: str) -> str:
+async def _maybe_url_image(img_url: str, target_path: str) -> str | None:
     if not isinstance(img_url, str):
         return None
 
@@ -86,25 +86,36 @@ async def _save_url_image_to_path(image_url: str, target_path: str) -> str:
     os.makedirs(os.path.dirname(target_path), exist_ok=True)
 
     try:
-        async with httpx.AsyncClient() as client:
+        async with httpx.AsyncClient(follow_redirects=True) as client:
             response = await client.get(image_url, timeout=10.0)
             response.raise_for_status()
 
             # Determine file extension from content type or URL after downloading
             if not os.path.splitext(target_path)[1]:
-                content_type = response.headers.get("content-type", "")
-                if not content_type.startswith("image/"):
+                content_type = response.headers.get("content-type", "").lower()
+
+                url_path = image_url.split("?")[0]
+                _, url_ext = os.path.splitext(url_path)
+                url_ext = url_ext.lower()
+
+                if url_ext in {".jpg", ".jpeg", ".png", ".webp", ".gif", ".bmp"}:
+                    ext = ".jpg" if url_ext == ".jpeg" else url_ext
+                elif content_type.startswith("image/"):
+                    if "jpeg" in content_type or "jpg" in content_type:
+                        ext = ".jpg"
+                    elif "png" in content_type:
+                        ext = ".png"
+                    elif "webp" in content_type:
+                        ext = ".webp"
+                    else:
+                        ext = ".jpg"  # Default to jpg
+                elif content_type == "application/octet-stream":
+                    # for octet-stream, if we couldn't get it from URL, default to jpg
+                    ext = ".jpg"
+                else:
                     raise ValueError(
                         f"URL does not point to an image. Content-Type: {content_type}"
                     )
-                if "jpeg" in content_type or "jpg" in content_type:
-                    ext = ".jpg"
-                elif "png" in content_type:
-                    ext = ".png"
-                elif "webp" in content_type:
-                    ext = ".webp"
-                else:
-                    ext = ".jpg"  # Default to jpg
                 target_path = f"{target_path}{ext}"
 
             with open(target_path, "wb") as f: