diff --git a/litellm/constants.py b/litellm/constants.py index 8d05115e76e..981f3cb215c 100644 --- a/litellm/constants.py +++ b/litellm/constants.py @@ -1093,6 +1093,7 @@ 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://inference.runanywhere.ai/v1", "https://gigachat.devices.sberbank.ru/api/v1", "https://api.openinfer.ai/v1", @@ -1180,6 +1181,7 @@ openai_compatible_providers: Final[list[str]] = [ "hostinger_router", "sail", "reka", + "linkrra", "streamlake", "wally", ] diff --git a/litellm/llms/openai_like/providers.json b/litellm/llms/openai_like/providers.json index f937e4d2f75..7bf472563c4 100644 --- a/litellm/llms/openai_like/providers.json +++ b/litellm/llms/openai_like/providers.json @@ -275,6 +275,12 @@ "/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"] + }, "refuseless": { "base_url": "https://api.refuseless.com/v1", "api_key_env": "REFUSELESS_API_KEY", diff --git a/litellm/provider_endpoints_support_backup.json b/litellm/provider_endpoints_support_backup.json index 00fee0a0d5a..8b1df373927 100644 --- a/litellm/provider_endpoints_support_backup.json +++ b/litellm/provider_endpoints_support_backup.json @@ -3682,6 +3682,24 @@ "generateContent": 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", diff --git a/litellm/proxy/public_endpoints/provider_create_fields.json b/litellm/proxy/public_endpoints/provider_create_fields.json index 5ded2fa5f5e..c4e7375ad05 100644 --- a/litellm/proxy/public_endpoints/provider_create_fields.json +++ b/litellm/proxy/public_endpoints/provider_create_fields.json @@ -3595,6 +3595,34 @@ ], "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": "WALLY", "provider_display_name": "Wally", diff --git a/litellm/types/utils.py b/litellm/types/utils.py index ca434154f68..e1517225f53 100644 --- a/litellm/types/utils.py +++ b/litellm/types/utils.py @@ -4783,6 +4783,7 @@ class LlmProviders(str, Enum): META = "meta" SAIL = "sail" REKA = "reka" + LINKRRA = "linkrra" STREAMLAKE = "streamlake" WALLY = "wally" LITELLM_AGENT = "litellm_agent" diff --git a/provider_endpoints_support.json b/provider_endpoints_support.json index 00fee0a0d5a..395d632689b 100644 --- a/provider_endpoints_support.json +++ b/provider_endpoints_support.json @@ -5250,6 +5250,24 @@ "generateContent": 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 + } + }, "wally": { "display_name": "Wally (`wally`)", "url": "https://docs.litellm.ai/docs/providers/wally", diff --git a/tests/unit/llms/openai_like/test_linkrra_provider.py b/tests/unit/llms/openai_like/test_linkrra_provider.py new file mode 100644 index 00000000000..dc0c49e2397 --- /dev/null +++ b/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"