Merge 97290ec4f8 into aeede53112

litellm/constants.py

@@ -1098,6 +1098,7 @@ openai_compatible_endpoints: Final[list[str]] = [
     "https://gigachat.devices.sberbank.ru/api/v1",
     "https://api.openinfer.ai/v1",
     "https://vanchin.streamlake.ai/api/gateway/v1/endpoints",
+    "https://api.uberllm.dev/v1",
 ]
 
 
@@ -1183,6 +1184,7 @@ openai_compatible_providers: Final[list[str]] = [
     "reka",
     "streamlake",
     "wally",
+    "uberllm",
 ]
 
 OPENAI_AUDIO_TRANSCRIPTION_PROVIDERS: Final = frozenset({"openai"} | frozenset(openai_compatible_providers))

litellm/llms/openai_like/providers.json

@@ -314,5 +314,15 @@
       "supports_function_calling": true,
       "supports_reasoning": true
     }
+  },
+  "uberllm": {
+    "base_url": "https://api.uberllm.dev/v1",
+    "api_key_env": "UBERLLM_API_KEY",
+    "api_base_env": "UBERLLM_API_BASE",
+    "provider_info": {
+      "supports_function_calling": true,
+      "supports_reasoning": true
+    },
+    "supported_endpoints": ["/v1/chat/completions", "/v1/messages"]
   }
 }

litellm/provider_endpoints_support_backup.json

@@ -4280,6 +4280,24 @@
         "generateContent": false
       }
     },
+    "uberllm": {
+      "display_name": "UberLLM (`uberllm`)",
+      "url": "https://docs.litellm.ai/docs/providers/uberllm",
+      "endpoints": {
+        "chat_completions": true,
+        "messages": true,
+        "responses": false,
+        "embeddings": false,
+        "image_generations": false,
+        "audio_transcriptions": false,
+        "audio_speech": false,
+        "moderations": false,
+        "batches": false,
+        "rerank": false,
+        "a2a": false,
+        "interactions": false
+      }
+    },
     "v0": {
       "display_name": "V0 (`v0`)",
       "url": "https://docs.litellm.ai/docs/providers/v0",

litellm/proxy/public_endpoints/provider_create_fields.json

@@ -4508,5 +4508,33 @@
       }
     ],
     "default_model_placeholder": "bigmodel/glm-5.3"
+  },
+  {
+    "provider": "UBERLLM",
+    "provider_display_name": "UberLLM",
+    "litellm_provider": "uberllm",
+    "credential_fields": [
+      {
+        "key": "api_base",
+        "label": "API Base",
+        "placeholder": "https://api.uberllm.dev/v1",
+        "tooltip": null,
+        "required": false,
+        "field_type": "text",
+        "options": null,
+        "default_value": null
+      },
+      {
+        "key": "api_key",
+        "label": "API Key",
+        "placeholder": null,
+        "tooltip": null,
+        "required": true,
+        "field_type": "password",
+        "options": null,
+        "default_value": null
+      }
+    ],
+    "default_model_placeholder": "uberllm/deepseek/deepseek-v4.1-flash"
   }
 ]

litellm/types/utils.py

@@ -4801,6 +4801,7 @@ class LlmProviders(str, Enum):
     BEDROCK_MANTLE = "bedrock_mantle"
     GDC = "gdc"
     KYMA = "kyma"
+    UBERLLM = "uberllm"
 
 
 # Create a set of all provider values for quick lookup

provider_endpoints_support.json

@@ -4280,6 +4280,24 @@
         "generateContent": false
       }
     },
+    "uberllm": {
+      "display_name": "UberLLM (`uberllm`)",
+      "url": "https://docs.litellm.ai/docs/providers/uberllm",
+      "endpoints": {
+        "chat_completions": true,
+        "messages": true,
+        "responses": false,
+        "embeddings": false,
+        "image_generations": false,
+        "audio_transcriptions": false,
+        "audio_speech": false,
+        "moderations": false,
+        "batches": false,
+        "rerank": false,
+        "a2a": false,
+        "interactions": false
+      }
+    },
     "v0": {
       "display_name": "V0 (`v0`)",
       "url": "https://docs.litellm.ai/docs/providers/v0",

tests/unit/llms/openai_like/test_uberllm_provider.py

@@ -0,0 +1,153 @@
+import json
+from pathlib import Path
+from typing import Final
+
+import pytest
+import respx
+
+import litellm
+from litellm.caching.llm_caching_handler import LLMClientCache
+
+
+def test_uberllm_provider_resolution(monkeypatch: pytest.MonkeyPatch):
+    from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
+
+    monkeypatch.setenv("UBERLLM_API_KEY", "uberllm-test-key")
+
+    model, provider, api_key, api_base = get_llm_provider(
+        model="uberllm/deepseek/deepseek-v4.1-flash",
+        custom_llm_provider=None,
+        api_base=None,
+        api_key=None,
+    )
+
+    assert model == "deepseek/deepseek-v4.1-flash"
+    assert provider == "uberllm"
+    assert api_key == "uberllm-test-key"
+    assert api_base == "https://api.uberllm.dev/v1"
+
+
+def test_uberllm_provider_keeps_explicit_credentials(monkeypatch: pytest.MonkeyPatch):
+    from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
+
+    monkeypatch.setenv("UBERLLM_API_KEY", "uberllm-env-key")
+
+    _, provider, api_key, api_base = get_llm_provider(
+        model="uberllm/deepseek/deepseek-v4.1-flash",
+        custom_llm_provider=None,
+        api_base="https://uberllm.internal.example/v1",
+        api_key="uberllm-explicit-key",
+    )
+
+    assert provider == "uberllm"
+    assert api_key == "uberllm-explicit-key"
+    assert api_base == "https://uberllm.internal.example/v1"
+
+
+def test_uberllm_is_available_in_add_model_form():
+    fields_path = Path(litellm.__file__).parent / "proxy" / "public_endpoints" / "provider_create_fields.json"
+    providers = json.loads(fields_path.read_text())
+    uberllm = next(provider for provider in providers if provider["litellm_provider"] == "uberllm")
+
+    assert uberllm["provider"] == "UBERLLM"
+    assert uberllm["provider_display_name"] == "UberLLM"
+    assert uberllm["default_model_placeholder"] == "uberllm/deepseek/deepseek-v4.1-flash"
+    assert {field["key"]: field["required"] for field in uberllm["credential_fields"]} == {
+        "api_base": False,
+        "api_key": True,
+    }
+
+
+def test_uberllm_supported_endpoints():
+    matrix_path = Path(litellm.__file__).parent / "provider_endpoints_support_backup.json"
+    providers = json.loads(matrix_path.read_text())["providers"]
+
+    assert providers["uberllm"]["endpoints"] == {
+        "chat_completions": True,
+        "messages": True,
+        "responses": False,
+        "embeddings": False,
+        "image_generations": False,
+        "audio_transcriptions": False,
+        "audio_speech": False,
+        "moderations": False,
+        "batches": False,
+        "rerank": False,
+        "a2a": False,
+        "interactions": False,
+    }
+
+
+@pytest.mark.parametrize("reasoning_effort", [None, "none", "high"])
+def test_uberllm_chat_completion_request(reasoning_effort: str | None):
+    effort_kwargs: Final = {} if reasoning_effort is None else {"reasoning_effort": reasoning_effort}
+    with respx.mock() as upstream:
+        route: Final = upstream.post("https://api.uberllm.dev/v1/chat/completions").respond(
+            200,
+            json={
+                "id": "chatcmpl_uberllm",
+                "object": "chat.completion",
+                "created": 1_793_330_000,
+                "model": "deepseek/deepseek-v4.1-flash",
+                "choices": [
+                    {
+                        "index": 0,
+                        "message": {"role": "assistant", "content": "Hello from UberLLM"},
+                        "finish_reason": "stop",
+                    }
+                ],
+                "usage": {"prompt_tokens": 4, "completion_tokens": 3, "total_tokens": 7},
+            },
+        )
+        response: Final = litellm.completion(
+            model="uberllm/deepseek/deepseek-v4.1-flash",
+            messages=[{"role": "user", "content": "Say hello"}],
+            api_key="uberllm-test-key",
+            **effort_kwargs,
+        )
+
+    request: Final = route.calls.last.request
+    body: Final = json.loads(request.content)
+    assert route.call_count == 1
+    assert str(request.url) == "https://api.uberllm.dev/v1/chat/completions"
+    assert body.get("reasoning_effort") == reasoning_effort
+    assert request.headers["authorization"] == "Bearer uberllm-test-key"
+    assert body["model"] == "deepseek/deepseek-v4.1-flash"
+    assert body["messages"] == [{"role": "user", "content": "Say hello"}]
+    assert response.choices[0].message.content == "Hello from UberLLM"
+
+
+@pytest.mark.asyncio
+async def test_uberllm_anthropic_messages_request(monkeypatch: pytest.MonkeyPatch):
+    monkeypatch.setattr(litellm, "disable_aiohttp_transport", True)
+    monkeypatch.setattr(litellm, "in_memory_llm_clients_cache", LLMClientCache())
+    with respx.mock() as upstream:
+        route: Final = upstream.post("https://api.uberllm.dev/v1/messages").respond(
+            200,
+            json={
+                "id": "msg_uberllm",
+                "type": "message",
+                "role": "assistant",
+                "model": "deepseek/deepseek-v4.1-flash",
+                "content": [{"type": "text", "text": "Hello from UberLLM"}],
+                "stop_reason": "end_turn",
+                "stop_sequence": None,
+                "usage": {"input_tokens": 4, "output_tokens": 3},
+            },
+        )
+        response: Final = await litellm.anthropic.messages.acreate(
+            model="uberllm/deepseek/deepseek-v4.1-flash",
+            messages=[{"role": "user", "content": "Say hello"}],
+            max_tokens=32,
+            api_key="uberllm-test-key",
+        )
+
+    request: Final = route.calls.last.request
+    body: Final = json.loads(request.content)
+    assert route.call_count == 1
+    assert str(request.url) == "https://api.uberllm.dev/v1/messages"
+    assert request.headers["authorization"] == "Bearer uberllm-test-key"
+    assert request.headers["anthropic-version"] == "2023-06-01"
+    assert body["model"] == "deepseek/deepseek-v4.1-flash"
+    assert body["messages"] == [{"role": "user", "content": "Say hello"}]
+    assert response["content"][0]["text"] == "Hello from UberLLM"