Merge 82f88de398 into da80552897

litellm/constants.py

@@ -1044,6 +1044,7 @@ openai_compatible_endpoints: Final[list[str]] = [
     "https://api.scx.ai/v1",
     "https://api.prisminference.com/v1",
     "https://api.reka.ai/v1",
+    "https://pool.linkrra.com/v1",
     "https://gigachat.devices.sberbank.ru/api/v1",
     "https://api.openinfer.ai/v1",
 ]
@@ -1124,6 +1125,7 @@ openai_compatible_providers: Final[list[str]] = [
     "prism",
     "sail",
     "reka",
+    "linkrra",
 ]
 
 OPENAI_AUDIO_TRANSCRIPTION_PROVIDERS: Final = frozenset({"openai"} | frozenset(openai_compatible_providers))

litellm/llms/openai_like/providers.json

@@ -249,5 +249,11 @@
       "/v1/responses",
       "/v1/messages"
     ]
+  },
+  "linkrra": {
+    "base_url": "https://pool.linkrra.com/v1",
+    "api_key_env": "LINKRRA_API_KEY",
+    "api_base_env": "LINKRRA_API_BASE",
+    "supported_endpoints": ["/v1/chat/completions", "/v1/messages"]
   }
 }

litellm/provider_endpoints_support_backup.json

@@ -2154,6 +2154,24 @@
         "interactions": false
       }
     },
+    "linkrra": {
+      "display_name": "Linkrra (`linkrra`)",
+      "url": "https://docs.litellm.ai/docs/providers/linkrra",
+      "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
+      }
+    },
     "scaleway": {
       "display_name": "Scaleway (`scaleway`)",
       "url": "https://docs.litellm.ai/docs/providers/scaleway",

litellm/proxy/public_endpoints/provider_create_fields.json

@@ -3287,6 +3287,34 @@
     ],
     "default_model_placeholder": "reka/reka-flash"
   },
+  {
+    "provider": "LINKRRA",
+    "provider_display_name": "Linkrra",
+    "litellm_provider": "linkrra",
+    "credential_fields": [
+      {
+        "key": "api_base",
+        "label": "API Base",
+        "placeholder": "https://pool.linkrra.com/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": "linkrra/qwen3-coder-30b"
+  },
   {
     "provider": "Sail",
     "provider_display_name": "Sail",

litellm/types/utils.py

@@ -4522,6 +4522,7 @@ class LlmProviders(str, Enum):
     META = "meta"
     SAIL = "sail"
     REKA = "reka"
+    LINKRRA = "linkrra"
     LITELLM_AGENT = "litellm_agent"
     CURSOR = "cursor"
     BEDROCK_MANTLE = "bedrock_mantle"

provider_endpoints_support.json

@@ -3223,6 +3223,24 @@
         "interactions": false
       }
     },
+    "linkrra": {
+      "display_name": "Linkrra (`linkrra`)",
+      "url": "https://docs.litellm.ai/docs/providers/linkrra",
+      "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
+      }
+    },
     "charity_engine": {
       "display_name": "Charity Engine (`charity_engine`)",
       "url": "https://docs.litellm.ai/docs/providers/charity_engine",

tests/unit/llms/openai_like/test_linkrra_provider.py

@@ -0,0 +1,205 @@
+import json
+from typing import Final
+
+import pytest
+import respx
+
+import litellm
+from litellm.caching.llm_caching_handler import LLMClientCache
+
+_API_BASE: Final = "https://pool.linkrra.com/v1"
+_CHAT_COMPLETION: Final = {
+    "id": "chatcmpl_linkrra",
+    "object": "chat.completion",
+    "created": 1_790_000_000,
+    "model": "qwen3-coder-30b",
+    "choices": [
+        {
+            "index": 0,
+            "message": {"role": "assistant", "content": "Hello from Linkrra"},
+            "finish_reason": "stop",
+        }
+    ],
+    "usage": {"prompt_tokens": 4, "completion_tokens": 3, "total_tokens": 7},
+}
+
+
+def test_linkrra_is_a_registered_provider():
+    assert litellm.LlmProviders.LINKRRA.value == "linkrra"
+    assert "linkrra" in litellm.provider_list
+    assert "linkrra" in litellm.constants.openai_compatible_providers
+
+
+def test_linkrra_provider_resolution(monkeypatch: pytest.MonkeyPatch):
+    from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
+
+    monkeypatch.delenv("LINKRRA_API_BASE", raising=False)
+    monkeypatch.setenv("LINKRRA_API_KEY", "linkrra-test-key")
+
+    model, provider, api_key, api_base = get_llm_provider(
+        model="linkrra/qwen3-coder-30b",
+        custom_llm_provider=None,
+        api_base=None,
+        api_key=None,
+    )
+
+    assert model == "qwen3-coder-30b"
+    assert provider == "linkrra"
+    assert api_key == "linkrra-test-key"
+    assert api_base == _API_BASE
+
+
+def test_linkrra_provider_keeps_explicit_credentials(monkeypatch: pytest.MonkeyPatch):
+    from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
+
+    monkeypatch.setenv("LINKRRA_API_KEY", "linkrra-env-key")
+    monkeypatch.setenv("LINKRRA_API_BASE", "https://linkrra.env.example/v1")
+
+    _, provider, api_key, api_base = get_llm_provider(
+        model="linkrra/qwen3-coder-30b",
+        custom_llm_provider=None,
+        api_base="https://linkrra.internal.example/v1",
+        api_key="linkrra-explicit-key",
+    )
+
+    assert provider == "linkrra"
+    assert api_key == "linkrra-explicit-key"
+    assert api_base == "https://linkrra.internal.example/v1"
+
+
+def test_linkrra_api_base_env_overrides_default(monkeypatch: pytest.MonkeyPatch):
+    from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
+
+    monkeypatch.setenv("LINKRRA_API_KEY", "linkrra-env-key")
+    monkeypatch.setenv("LINKRRA_API_BASE", "https://linkrra.env.example/v1")
+
+    _, provider, _, api_base = get_llm_provider(
+        model="linkrra/qwen3-coder-30b",
+        custom_llm_provider=None,
+        api_base=None,
+        api_key=None,
+    )
+
+    assert provider == "linkrra"
+    assert api_base == "https://linkrra.env.example/v1"
+
+
+def test_linkrra_api_base_autodetects_provider(monkeypatch: pytest.MonkeyPatch):
+    from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
+
+    monkeypatch.setenv("LINKRRA_API_KEY", "linkrra-env-key")
+
+    model, provider, api_key, api_base = get_llm_provider(
+        model="qwen3-coder-30b",
+        custom_llm_provider=None,
+        api_base=_API_BASE,
+        api_key=None,
+    )
+
+    assert model == "qwen3-coder-30b"
+    assert provider == "linkrra"
+    assert api_key == "linkrra-env-key"
+    assert api_base == _API_BASE
+
+
+@pytest.mark.asyncio
+async def test_linkrra_is_served_by_the_add_model_form_endpoint():
+    from litellm.proxy.public_endpoints.public_endpoints import get_provider_fields
+
+    fields: Final = await get_provider_fields()
+    linkrra: Final = next(provider for provider in fields if provider.litellm_provider == "linkrra")
+
+    assert linkrra.provider_display_name == "Linkrra"
+    assert linkrra.default_model_placeholder == "linkrra/qwen3-coder-30b"
+    assert {field.key: field.required for field in linkrra.credential_fields} == {
+        "api_base": False,
+        "api_key": True,
+    }
+
+
+@pytest.mark.asyncio
+async def test_linkrra_is_listed_only_under_the_endpoints_it_serves():
+    from litellm.proxy.public_endpoints.public_endpoints import get_supported_endpoints
+
+    supported: Final = await get_supported_endpoints()
+    listed_under: Final = {
+        entry["key"]
+        for entry in supported.model_dump()["endpoints"]
+        if "linkrra" in {provider["slug"] for provider in entry["providers"]}
+    }
+
+    assert listed_under == {"chat_completions", "messages"}
+
+
+def test_linkrra_chat_completion_request():
+    with respx.mock() as upstream:
+        route: Final = upstream.post(f"{_API_BASE}/chat/completions").respond(200, json=_CHAT_COMPLETION)
+        response: Final = litellm.completion(
+            model="linkrra/qwen3-coder-30b",
+            messages=[{"role": "user", "content": "Say hello"}],
+            api_key="linkrra-test-key",
+        )
+
+    request: Final = route.calls.last.request
+    body: Final = json.loads(request.content)
+    assert route.call_count == 1
+    assert str(request.url) == f"{_API_BASE}/chat/completions"
+    assert request.headers["authorization"] == "Bearer linkrra-test-key"
+    assert body["model"] == "qwen3-coder-30b"
+    assert body["messages"] == [{"role": "user", "content": "Say hello"}]
+    assert response.choices[0].message.content == "Hello from Linkrra"
+
+
+def test_linkrra_responses_request_is_bridged_to_chat_completions():
+    with respx.mock() as upstream:
+        route: Final = upstream.post(f"{_API_BASE}/chat/completions").respond(200, json=_CHAT_COMPLETION)
+        response: Final = litellm.responses(
+            model="linkrra/qwen3-coder-30b",
+            input="Say hello",
+            api_key="linkrra-test-key",
+        )
+
+    request: Final = route.calls.last.request
+    body: Final = json.loads(request.content)
+    assert route.call_count == 1
+    assert str(request.url) == f"{_API_BASE}/chat/completions"
+    assert request.headers["authorization"] == "Bearer linkrra-test-key"
+    assert body["model"] == "qwen3-coder-30b"
+    assert body["messages"] == [{"role": "user", "content": "Say hello"}]
+    assert response.output[0].content[0].text == "Hello from Linkrra"
+
+
+@pytest.mark.asyncio
+async def test_linkrra_anthropic_messages_request_is_sent_natively(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(f"{_API_BASE}/messages").respond(
+            200,
+            json={
+                "id": "msg_linkrra",
+                "type": "message",
+                "role": "assistant",
+                "model": "qwen3-coder-30b",
+                "content": [{"type": "text", "text": "Hello from Linkrra"}],
+                "stop_reason": "end_turn",
+                "stop_sequence": None,
+                "usage": {"input_tokens": 4, "output_tokens": 3},
+            },
+        )
+        response: Final = await litellm.anthropic.messages.acreate(
+            model="linkrra/qwen3-coder-30b",
+            messages=[{"role": "user", "content": "Say hello"}],
+            max_tokens=32,
+            api_key="linkrra-test-key",
+        )
+
+    request: Final = route.calls.last.request
+    body: Final = json.loads(request.content)
+    assert route.call_count == 1
+    assert str(request.url) == f"{_API_BASE}/messages"
+    assert request.headers["authorization"] == "Bearer linkrra-test-key"
+    assert body["model"] == "qwen3-coder-30b"
+    assert body["messages"] == [{"role": "user", "content": "Say hello"}]
+    assert body["max_tokens"] == 32
+    assert response["content"][0]["text"] == "Hello from Linkrra"