feat(greenpt): add flagship and rerank support

litellm/llms/greenpt/__init__.py

@@ -0,0 +1 @@
+

litellm/llms/greenpt/rerank/__init__.py

@@ -0,0 +1 @@
+

litellm/llms/greenpt/rerank/transformation.py

@@ -0,0 +1,42 @@
+from typing import Dict
+
+from litellm.llms.hosted_vllm.rerank.transformation import HostedVLLMRerankConfig
+from litellm.secret_managers.main import get_secret_str
+
+
+class GreenPTRerankConfig(HostedVLLMRerankConfig):
+    def get_supported_cohere_rerank_params(self, model: str) -> list:
+        return ["query", "documents", "top_n", "return_documents"]
+
+    def validate_environment(
+        self,
+        headers: dict,
+        model: str,
+        api_key: str | None = None,
+        optional_params: dict | None = None,
+    ) -> dict:
+        api_key = api_key or get_secret_str("GREENPT_API_KEY")
+        if api_key is None:
+            raise ValueError("GreenPT API key is required. Set GREENPT_API_KEY.")
+        return super().validate_environment(
+            headers=headers,
+            model=model,
+            api_key=api_key,
+            optional_params=optional_params,
+        )
+
+    def transform_rerank_request(
+        self,
+        model: str,
+        optional_rerank_params: Dict,
+        headers: dict,
+        litellm_params: dict | None = None,
+    ) -> dict:
+        request = super().transform_rerank_request(
+            model=model,
+            optional_rerank_params=optional_rerank_params,
+            headers=headers,
+            litellm_params=litellm_params,
+        )
+        request.setdefault("top_n", len(request["documents"]))
+        return request

litellm/rerank_api/main.py

@@ -31,7 +31,7 @@ async def arerank(
     query: str,
     documents: List[Union[str, Dict[str, Any]]],
     custom_llm_provider: (
-        Literal["cohere", "together_ai", "deepinfra", "fireworks_ai", "voyage", "watsonx"] | None
+        Literal["cohere", "together_ai", "deepinfra", "fireworks_ai", "voyage", "watsonx", "greenpt"] | None
     ) = None,
     top_n: int | None = None,
     rank_fields: List[str] | None = None,
@@ -89,6 +89,7 @@ def rerank(
             "fireworks_ai",
             "voyage",
             "watsonx",
+            "greenpt",
         ]
         | None
     ) = None,

litellm/utils.py

@@ -8008,6 +8008,10 @@ class ProviderConfigManager:
             )
 
             return DashScopeRerankConfig()
+        elif litellm.LlmProviders.GREENPT == provider:
+            from litellm.llms.greenpt.rerank.transformation import GreenPTRerankConfig
+
+            return GreenPTRerankConfig()
         return litellm.CohereRerankConfig()
 
     @staticmethod

model_prices_and_context_window.json

@@ -46295,6 +46295,14 @@
         "output_cost_per_token": 0,
         "source": "https://docs.greenpt.ai/pricing"
     },
+    "greenpt/green-rerank": {
+        "input_cost_per_token": 1.36524e-07,
+        "litellm_provider": "greenpt",
+        "max_input_tokens": 32768,
+        "max_tokens": 32768,
+        "mode": "rerank",
+        "source": "https://docs.greenpt.ai/pricing"
+    },
     "greenpt/green-l": {
         "input_cost_per_token": 2.84425e-07,
         "litellm_provider": "greenpt",
@@ -46370,6 +46378,19 @@
         "supports_tool_choice": true,
         "supports_vision": true
     },
+    "greenpt/kimi-k2.7-code": {
+        "input_cost_per_token": 9.44291e-07,
+        "litellm_provider": "greenpt",
+        "max_input_tokens": 262144,
+        "max_output_tokens": 262144,
+        "max_tokens": 262144,
+        "mode": "chat",
+        "output_cost_per_token": 4.380145e-06,
+        "source": "https://docs.greenpt.ai/pricing",
+        "supports_function_calling": true,
+        "supports_reasoning": true,
+        "supports_tool_choice": true
+    },
     "greenpt/minimax-m2.5": {
         "input_cost_per_token": 1.93409e-07,
         "litellm_provider": "greenpt",
@@ -46383,6 +46404,16 @@
         "supports_reasoning": true,
         "supports_tool_choice": true
     },
+    "greenpt/qwen3-embedding-8b": {
+        "input_cost_per_token": 2.84425e-07,
+        "litellm_provider": "greenpt",
+        "max_input_tokens": 32768,
+        "max_output_tokens": 4096,
+        "max_tokens": 32768,
+        "mode": "embedding",
+        "output_cost_per_token": 0,
+        "source": "https://docs.greenpt.ai/pricing"
+    },
     "fallback_generalizations": {
         "rules": [
             {

provider_endpoints_support.json

@@ -2955,7 +2955,7 @@
         "audio_speech": false,
         "moderations": false,
         "batches": false,
-        "rerank": false,
+        "rerank": true,
         "a2a": false
       }
     }

tests/litellm/llms/openai_like/test_greenpt_provider.py

@@ -5,6 +5,8 @@ sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), "../.
 
 from litellm.llms.openai_like.dynamic_config import create_config_class
 from litellm.llms.openai_like.json_loader import JSONProviderRegistry
+from litellm.types.utils import LlmProviders
+from litellm.utils import ProviderConfigManager
 
 GREENPT_BASE_URL = "https://api.greenpt.ai/v1"
 
@@ -69,3 +71,28 @@ def test_greenpt_provider_resolution():
     assert model == "gemma4"
     assert provider == "greenpt"
     assert api_base == GREENPT_BASE_URL
+
+
+def test_greenpt_rerank_config():
+    config = ProviderConfigManager.get_provider_rerank_config(
+        model="green-rerank",
+        provider=LlmProviders.GREENPT,
+        api_base=GREENPT_BASE_URL,
+        present_version_params=[],
+    )
+    assert config.get_complete_url(GREENPT_BASE_URL, "green-rerank") == (f"{GREENPT_BASE_URL}/rerank")
+    assert config.transform_rerank_request(
+        model="green-rerank",
+        optional_rerank_params={
+            "query": "low-carbon inference",
+            "documents": ["GreenPT", "Other provider"],
+            "return_documents": False,
+        },
+        headers={},
+    ) == {
+        "model": "green-rerank",
+        "query": "low-carbon inference",
+        "documents": ["GreenPT", "Other provider"],
+        "top_n": 2,
+        "return_documents": False,
+    }