feat: Add comprehensive GPT-5 series model support

- Add GPT-5, GPT-5-mini, and GPT-5-nano models to unified configuration
- Implement proper thinking mode support via dynamic capability checking
- Add OpenAI provider model enumeration methods for registry integration
- Update tests to cover all GPT-5 models and their aliases
- Fix critical bug where thinking mode was hardcoded instead of using model capabilities

Breaking Changes:
- None (backward compatible)

New Models Available:
- gpt-5 (400K context, 128K output, reasoning support)
- gpt-5-mini (400K context, 128K output, efficient variant)
- gpt-5-nano (400K context, fastest/cheapest variant)

Aliases:
- gpt5, gpt5-mini, gpt5mini, gpt5-nano, gpt5nano, nano

All models support:
- Extended thinking mode (reasoning tokens)
- Vision capabilities
- JSON mode
- Function calling

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>

conf/custom_models.json

@@ -228,6 +228,48 @@
       "temperature_constraint": "fixed",
       "description": "OpenAI's o4-mini model - optimized for shorter contexts with rapid reasoning and vision"
     },
+    {
+      "model_name": "gpt-5",
+      "aliases": ["gpt5", "gpt-5"],
+      "context_window": 400000,
+      "max_output_tokens": 128000,
+      "supports_extended_thinking": true,
+      "supports_json_mode": true,
+      "supports_function_calling": true,
+      "supports_images": true,
+      "max_image_size_mb": 20.0,
+      "supports_temperature": true,
+      "temperature_constraint": "fixed",
+      "description": "GPT-5 (400K context, 128K output) - Advanced model with reasoning support"
+    },
+    {
+      "model_name": "gpt-5-mini",
+      "aliases": ["gpt5-mini", "gpt5mini", "mini"],
+      "context_window": 400000,
+      "max_output_tokens": 128000,
+      "supports_extended_thinking": true,
+      "supports_json_mode": true,
+      "supports_function_calling": true,
+      "supports_images": true,
+      "max_image_size_mb": 20.0,
+      "supports_temperature": true,
+      "temperature_constraint": "fixed",
+      "description": "GPT-5-mini (400K context, 128K output) - Efficient variant with reasoning support"
+    },
+    {
+      "model_name": "gpt-5-nano",
+      "aliases": ["gpt5nano", "gpt5-nano", "nano"],
+      "context_window": 400000,
+      "max_output_tokens": 128000,
+      "supports_extended_thinking": true,
+      "supports_json_mode": true,
+      "supports_function_calling": true,
+      "supports_images": true,
+      "max_image_size_mb": 20.0,
+      "supports_temperature": true,
+      "temperature_constraint": "fixed",
+      "description": "GPT-5 nano (400K context) - Fastest, cheapest version of GPT-5 for summarization and classification tasks"
+    },
     {
       "model_name": "llama3.2",
       "aliases": ["local-llama", "local", "llama3.2", "ollama-llama"],

providers/openai_provider.py

@@ -259,12 +259,10 @@ class OpenAIModelProvider(OpenAICompatibleProvider):
 
     def supports_thinking_mode(self, model_name: str) -> bool:
         """Check if the model supports extended thinking mode."""
-        # GPT-5 models support reasoning tokens (extended thinking)
-        resolved_name = self._resolve_model_name(model_name)
-        if resolved_name in ["gpt-5", "gpt-5-mini"]:
-            return True
-        # O3 models don't support extended thinking yet
-        return False
+        try:
+            return self.get_capabilities(model_name).supports_extended_thinking
+        except ValueError:
+            return False
 
     def get_preferred_model(self, category: "ToolModelCategory", allowed_models: list[str]) -> Optional[str]:
         """Get OpenAI's preferred model for a given category from allowed models.
@@ -303,3 +301,19 @@ class OpenAIModelProvider(OpenAICompatibleProvider):
             # Prefer balanced performance/cost models
             preferred = find_first(["gpt-5", "gpt-5-mini", "o4-mini", "o3-mini"])
             return preferred if preferred else allowed_models[0]
+
+    def get_model_configurations(self) -> dict[str, ModelCapabilities]:
+        """Get model configurations supported by this provider.
+
+        Returns:
+            Dict mapping model names to their ModelCapabilities
+        """
+        return self.SUPPORTED_MODELS.copy()
+
+    def get_all_model_aliases(self) -> dict[str, list[str]]:
+        """Get all model aliases supported by this provider.
+
+        Returns:
+            Dict mapping model names to their alias lists
+        """
+        return {model_name: caps.aliases for model_name, caps in self.SUPPORTED_MODELS.items()}

tests/test_openai_provider.py

@@ -253,14 +253,21 @@ class TestOpenAIProvider:
         assert call_kwargs["model"] == "o3-mini"  # Should be unchanged
 
     def test_supports_thinking_mode(self):
-        """Test thinking mode support."""
+        """Test thinking mode support based on model capabilities."""
         provider = OpenAIModelProvider("test-key")
 
-        # GPT-5 models support thinking mode (reasoning tokens)
+        # GPT-5 models support thinking mode (reasoning tokens) - all variants
         assert provider.supports_thinking_mode("gpt-5") is True
         assert provider.supports_thinking_mode("gpt-5-mini") is True
-        assert provider.supports_thinking_mode("gpt5") is True  # Test with alias
-        assert provider.supports_thinking_mode("gpt5mini") is True  # Test with alias
+        assert provider.supports_thinking_mode("gpt-5-nano") is True  # Now included
+
+        # Test GPT-5 aliases
+        assert provider.supports_thinking_mode("gpt5") is True
+        assert provider.supports_thinking_mode("gpt5-mini") is True
+        assert provider.supports_thinking_mode("gpt5mini") is True
+        assert provider.supports_thinking_mode("gpt5-nano") is True
+        assert provider.supports_thinking_mode("gpt5nano") is True
+        assert provider.supports_thinking_mode("nano") is True  # New alias for gpt-5-nano
 
         # O3/O4 models don't support thinking mode
         assert provider.supports_thinking_mode("o3") is False
@@ -270,6 +277,9 @@ class TestOpenAIProvider:
             provider.supports_thinking_mode("mini") is True
         )  # "mini" now resolves to gpt-5-mini which supports thinking
 
+        # Test invalid model name
+        assert provider.supports_thinking_mode("invalid-model") is False
+
     @patch("providers.openai_compatible.OpenAI")
     def test_o3_pro_routes_to_responses_endpoint(self, mock_openai_class):
         """Test that o3-pro model routes to the /v1/responses endpoint (mock test)."""