fix(httpx): set custom timeout for all POST requests

src/rotator_library/providers/antigravity_provider.py

@@ -3703,7 +3703,10 @@ class AntigravityProvider(AntigravityAuthBase, ProviderInterface):
         file_logger: Optional[AntigravityFileLogger] = None,
     ) -> litellm.ModelResponse:
         """Handle non-streaming completion."""
-        response = await client.post(url, headers=headers, json=payload, timeout=600.0)
+        response = await client.post(
+            url, headers=headers, json=payload,
+            timeout=httpx.Timeout(connect=30.0, read=120.0, write=120.0, pool=120.0)
+        )
         response.raise_for_status()
 
         data = response.json()
@@ -3736,7 +3739,8 @@ class AntigravityProvider(AntigravityAuthBase, ProviderInterface):
         }
 
         async with client.stream(
-            "POST", url, headers=headers, json=payload, timeout=600.0
+            "POST", url, headers=headers, json=payload,
+            timeout=httpx.Timeout(connect=30.0, read=120.0, write=120.0, pool=120.0)
         ) as response:
             if response.status_code >= 400:
                 # Read error body for raise_for_status to include in exception

src/rotator_library/providers/gemini_cli_provider.py

@@ -1965,7 +1965,7 @@ class GeminiCliProvider(GeminiAuthBase, ProviderInterface):
                         headers=final_headers,
                         json=request_payload,
                         params={"alt": "sse"},
-                        timeout=600,
+                        timeout=httpx.Timeout(connect=30.0, read=120.0, write=120.0, pool=120.0),
                     ) as response:
                         # Read and log error body before raise_for_status for better debugging
                         if response.status_code >= 400:

src/rotator_library/providers/iflow_provider.py

@@ -492,7 +492,10 @@ class IFlowProvider(IFlowAuthBase, ProviderInterface):
             file_logger.log_request(payload)
             lib_logger.debug(f"iFlow Request URL: {url}")
 
-            return client.stream("POST", url, headers=headers, json=payload, timeout=600)
+            return client.stream(
+                "POST", url, headers=headers, json=payload,
+                timeout=httpx.Timeout(connect=30.0, read=120.0, write=120.0, pool=120.0)
+            )
 
         async def stream_handler(response_stream, attempt=1):
             """Handles the streaming response and converts chunks."""

src/rotator_library/providers/qwen_code_provider.py

@@ -503,7 +503,10 @@ class QwenCodeProvider(QwenAuthBase, ProviderInterface):
             file_logger.log_request(payload)
             lib_logger.debug(f"Qwen Code Request URL: {url}")
 
-            return client.stream("POST", url, headers=headers, json=payload, timeout=600)
+            return client.stream(
+                "POST", url, headers=headers, json=payload,
+                timeout=httpx.Timeout(connect=30.0, read=120.0, write=120.0, pool=120.0)
+            )
 
         async def stream_handler(response_stream, attempt=1):
             """Handles the streaming response and converts chunks."""