feat(openinfer): add OpenInfer as an OpenAI-compatible provider

litellm/constants.py

@@ -728,6 +728,7 @@
     "https://api.libertai.io/v1",
     "https://pinstripes.io/v1",
     "https://api.meta.ai/v1",
+    "https://api.openinfer.ai/v1",
 ]
 
 
@@ -793,6 +794,7 @@
     "docker_model_runner",
     "ragflow",
     "pinstripes",  # Pinstripes - JSON-configured provider
+    "openinfer",  # OpenInfer - JSON-configured provider
     "darkbloom",
     "meta",  # Meta Model API (Muse Spark) - JSON-configured provider
 ]

litellm/litellm_core_utils/get_llm_provider_logic.py

@@ -346,6 +346,9 @@ def get_llm_provider(
                     elif endpoint == "https://pinstripes.io/v1":
                         custom_llm_provider = "pinstripes"
                         dynamic_api_key = get_secret_str("PINSTRIPES_API_KEY")
+                    elif endpoint == "https://api.openinfer.ai/v1":
+                        custom_llm_provider = "openinfer"
+                        dynamic_api_key = get_secret_str("OPENINFER_API_KEY")
                     elif endpoint == "https://api.meta.ai/v1":
                         custom_llm_provider = "meta"
                         dynamic_api_key = get_secret_str("META_API_KEY")

litellm/llms/openai_like/providers.json

@@ -183,5 +183,15 @@
       "max_completion_tokens": "max_tokens"
     },
     "supported_endpoints": ["/v1/chat/completions", "/v1/responses", "/v1/embeddings"]
+  },
+  "openinfer": {
+    "base_url": "https://api.openinfer.ai/v1",
+    "api_key_env": "OPENINFER_API_KEY",
+    "api_base_env": "OPENINFER_API_BASE",
+    "base_class": "openai_gpt",
+    "param_mappings": {
+      "max_completion_tokens": "max_tokens"
+    },
+    "supported_endpoints": ["/v1/chat/completions"]
   }
 }

litellm/model_prices_and_context_window_backup.json

@@ -46190,5 +46190,16 @@
                 }
             }
         ]
+    },
+    "openinfer/llama-3.1-8b-instruct": {
+        "max_tokens": 8192,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 8192,
+        "input_cost_per_token": 0.0,
+        "output_cost_per_token": 0.0,
+        "litellm_provider": "openinfer",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "source": "https://api.openinfer.ai/v1/models"
     }
 }

litellm/types/utils.py

@@ -3508,6 +3508,7 @@ class LlmProviders(str, Enum):
     TENSORMESH = "tensormesh"
     LIBERTAI = "libertai"
     PINSTRIPES = "pinstripes"
+    OPENINFER = "openinfer"
     DARKBLOOM = "darkbloom"
     META = "meta"
     LITELLM_AGENT = "litellm_agent"

model_prices_and_context_window.json

@@ -46312,5 +46312,16 @@
                 }
             }
         ]
+    },
+    "openinfer/llama-3.1-8b-instruct": {
+        "max_tokens": 8192,
+        "max_input_tokens": 128000,
+        "max_output_tokens": 8192,
+        "input_cost_per_token": 0.0,
+        "output_cost_per_token": 0.0,
+        "litellm_provider": "openinfer",
+        "mode": "chat",
+        "supports_function_calling": true,
+        "source": "https://api.openinfer.ai/v1/models"
     }
 }

provider_endpoints_support.json

@@ -2018,6 +2018,23 @@
         "a2a": false
       }
     },
+    "openinfer": {
+      "display_name": "OpenInfer (`openinfer`)",
+      "url": "https://openinfer.io",
+      "endpoints": {
+        "chat_completions": true,
+        "messages": false,
+        "responses": false,
+        "embeddings": false,
+        "image_generations": false,
+        "audio_transcriptions": false,
+        "audio_speech": false,
+        "moderations": false,
+        "batches": false,
+        "rerank": false,
+        "a2a": false
+      }
+    },
     "poe": {
       "display_name": "Poe (`poe`)",
       "endpoints": {

tests/test_litellm/llms/openai_like/test_openinfer_provider.py

@@ -0,0 +1,91 @@
+"""
+Tests for the OpenInfer LLM provider configuration and integration.
+"""
+
+import litellm
+
+
+class TestOpenInferProviderConfig:
+    def test_openinfer_in_provider_list(self):
+        from litellm import LlmProviders
+
+        assert LlmProviders.OPENINFER.value == "openinfer"
+        assert "openinfer" in litellm.provider_list
+
+    def test_openinfer_json_config(self):
+        from litellm.llms.openai_like.json_loader import JSONProviderRegistry
+
+        provider = JSONProviderRegistry.get("openinfer")
+        assert provider is not None
+        assert provider.base_url == "https://api.openinfer.ai/v1"
+        assert provider.api_key_env == "OPENINFER_API_KEY"
+        assert provider.api_base_env == "OPENINFER_API_BASE"
+        assert not JSONProviderRegistry.supports_responses_api("openinfer")
+
+    def test_openinfer_in_openai_compatible_providers(self):
+        from litellm.constants import openai_compatible_providers
+
+        assert "openinfer" in openai_compatible_providers
+
+    def test_provider_prefixed_model_routes_to_openinfer(self):
+        from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
+
+        model, provider, api_key, api_base = get_llm_provider(
+            model="openinfer/llama-3.1-8b-instruct",
+            custom_llm_provider=None,
+            api_base=None,
+            api_key="sk-test",
+        )
+
+        assert model == "llama-3.1-8b-instruct"
+        assert provider == "openinfer"
+        assert api_key == "sk-test"
+        assert api_base == "https://api.openinfer.ai/v1"
+
+    def test_api_key_and_base_resolved_from_env(self, monkeypatch):
+        from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
+
+        monkeypatch.setenv("OPENINFER_API_KEY", "sk-env-key")
+        monkeypatch.setenv("OPENINFER_API_BASE", "https://proxy.internal/v1")
+
+        _, provider, api_key, api_base = get_llm_provider(
+            model="openinfer/llama-3.1-8b-instruct",
+            custom_llm_provider=None,
+            api_base=None,
+            api_key=None,
+        )
+
+        assert provider == "openinfer"
+        assert api_key == "sk-env-key"
+        assert api_base == "https://proxy.internal/v1"
+
+    def test_url_autodetection_from_api_base(self, monkeypatch):
+        from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
+
+        monkeypatch.setenv("OPENINFER_API_KEY", "sk-env-key")
+
+        _, provider, api_key, api_base = get_llm_provider(
+            model="llama-3.1-8b-instruct",
+            custom_llm_provider=None,
+            api_base="https://api.openinfer.ai/v1",
+            api_key=None,
+        )
+
+        assert provider == "openinfer"
+        assert api_key == "sk-env-key"
+
+    def test_chat_completions_url(self):
+        config = litellm.ProviderConfigManager.get_provider_chat_config(
+            model="llama-3.1-8b-instruct", provider=litellm.LlmProviders.OPENINFER
+        )
+        assert config is not None
+        assert (
+            config.get_complete_url(
+                api_base=None,
+                api_key="sk-test",
+                model="llama-3.1-8b-instruct",
+                optional_params={},
+                litellm_params={},
+            )
+            == "https://api.openinfer.ai/v1/chat/completions"
+        )