From b46f17faf5d56ce853fff94d0daa5ffa0d2fb428 Mon Sep 17 00:00:00 2001 From: mateo-berri <277851410+mateo-berri@users.noreply.github.com> Date: Tue, 25 Aug 2026 10:18:55 -0700 Subject: [PATCH 1/2] fix(together_ai): default endpoints to api.together.ai instead of api.together.xyz Together AI moved its canonical API host from api.together.xyz to api.together.ai. Default the provider api_base and the rerank handler to the new host, make rerank honor api_base and TOGETHER_AI_API_BASE like chat already does, map both hosts to together_ai when passed as api_base, and delete the dead models/info fetch in factory.py. --- basedpyright-code-budget.json | 8 +-- litellm/constants.py | 1 + .../get_llm_provider_logic.py | 10 ++- .../prompt_templates/factory.py | 43 ------------ litellm/llms/together_ai/rerank/handler.py | 12 +++- litellm/rerank_api/main.py | 3 + ruff-strict-budget.json | 6 +- .../test_get_llm_provider_endpoint_match.py | 39 +++++++++++ tests/test_litellm/rerank_api/test_main.py | 67 +++++++++++++++++++ type-discipline-budget.json | 2 +- 10 files changed, 136 insertions(+), 55 deletions(-) diff --git a/basedpyright-code-budget.json b/basedpyright-code-budget.json index 664e1669834..f4d4e25859a 100644 --- a/basedpyright-code-budget.json +++ b/basedpyright-code-budget.json @@ -1,6 +1,6 @@ { "reportAny": { - "limit": 19955 + "limit": 19949 }, "reportArgumentType": { "limit": 2566 @@ -54,7 +54,7 @@ "limit": 0 }, "reportMissingParameterType": { - "limit": 5663 + "limit": 5661 }, "reportMissingTypeArgument": { "limit": 15555 @@ -105,10 +105,10 @@ "limit": 109 }, "reportUnknownMemberType": { - "limit": 39011 + "limit": 39009 }, "reportUnknownParameterType": { - "limit": 19885 + "limit": 19883 }, "reportUnknownVariableType": { "limit": 30569 diff --git a/litellm/constants.py b/litellm/constants.py index 0a1ada3bab2..78aba30f9c0 100644 --- a/litellm/constants.py +++ b/litellm/constants.py @@ -750,6 +750,7 @@ openai_compatible_endpoints: Final[list] = [ "api.groq.com/openai/v1", "https://integrate.api.nvidia.com/v1", "api.deepseek.com/v1", + "api.together.ai/v1", "api.together.xyz/v1", "app.empower.dev/api/v1", "https://api.friendli.ai/serverless/v1", diff --git a/litellm/litellm_core_utils/get_llm_provider_logic.py b/litellm/litellm_core_utils/get_llm_provider_logic.py index e674fc37673..d2d82064c47 100644 --- a/litellm/litellm_core_utils/get_llm_provider_logic.py +++ b/litellm/litellm_core_utils/get_llm_provider_logic.py @@ -272,6 +272,14 @@ def get_llm_provider( elif endpoint == "api.deepseek.com/v1": custom_llm_provider = "deepseek" dynamic_api_key = get_secret_str("DEEPSEEK_API_KEY") + elif endpoint == "api.together.ai/v1" or endpoint == "api.together.xyz/v1": + custom_llm_provider = "together_ai" + dynamic_api_key = ( + get_secret_str("TOGETHER_API_KEY") + or get_secret_str("TOGETHER_AI_API_KEY") + or get_secret_str("TOGETHERAI_API_KEY") + or get_secret_str("TOGETHER_AI_TOKEN") + ) elif endpoint == "ollama.com": custom_llm_provider = "ollama" dynamic_api_key = get_secret_str("OLLAMA_API_KEY") @@ -707,7 +715,7 @@ def _get_openai_compatible_provider_info( dynamic_api_key, ) = litellm.ZAIChatConfig()._get_openai_compatible_provider_info(api_base, api_key) elif custom_llm_provider == "together_ai": - api_base = api_base or get_secret_str("TOGETHER_AI_API_BASE") or "https://api.together.xyz/v1" + api_base = api_base or get_secret_str("TOGETHER_AI_API_BASE") or "https://api.together.ai/v1" dynamic_api_key = api_key or ( get_secret_str("TOGETHER_API_KEY") or get_secret_str("TOGETHER_AI_API_KEY") diff --git a/litellm/litellm_core_utils/prompt_templates/factory.py b/litellm/litellm_core_utils/prompt_templates/factory.py index 826a890eca9..86cfbf70255 100644 --- a/litellm/litellm_core_utils/prompt_templates/factory.py +++ b/litellm/litellm_core_utils/prompt_templates/factory.py @@ -643,49 +643,6 @@ def claude_2_1_pt( return prompt -### TOGETHER AI - - -def get_model_info(token, model): - try: - headers: Final = {"Authorization": f"Bearer {token}"} - client: Final = HTTPHandler(concurrent_limit=1) - response: Final = client.get("https://api.together.xyz/models/info", headers=headers) - if response.status_code == 200: - model_info: Final = response.json() - for m in model_info: - if m["name"].lower().strip() == model.strip(): - return m["config"].get("prompt_format", None), m["config"].get("chat_template", None) - return None, None - else: - return None, None - except Exception: # safely fail a prompt template request - return None, None - - -## OLD TOGETHER AI FLOW -# def format_prompt_togetherai(messages, prompt_format, chat_template): -# if prompt_format is None: -# return default_pt(messages) - -# human_prompt, assistant_prompt = prompt_format.split("{prompt}") - -# if chat_template is not None: -# prompt = hf_chat_template( -# model=None, messages=messages, chat_template=chat_template -# ) -# elif prompt_format is not None: -# prompt = custom_prompt( -# role_dict={}, -# messages=messages, -# initial_prompt_value=human_prompt, -# final_prompt_value=assistant_prompt, -# ) -# else: -# prompt = default_pt(messages) -# return prompt - - ### IBM Granite diff --git a/litellm/llms/together_ai/rerank/handler.py b/litellm/llms/together_ai/rerank/handler.py index 10246451a9d..8407018b898 100644 --- a/litellm/llms/together_ai/rerank/handler.py +++ b/litellm/llms/together_ai/rerank/handler.py @@ -16,11 +16,16 @@ from litellm.llms.together_ai.rerank.transformation import TogetherAIRerankConfi from litellm.types.rerank import RerankRequest, RerankResponse +def _rerank_url(api_base: str) -> str: + return f"{api_base.rstrip('/')}/rerank" + + class TogetherAIRerank(BaseLLM): def rerank( self, model: str, api_key: str, + api_base: str, query: str, documents: list[str | dict[str, Any]], top_n: int | None = None, @@ -46,10 +51,10 @@ class TogetherAIRerank(BaseLLM): raise ValueError("TogetherAI does not support max_chunks_per_doc") if _is_async: - return self.async_rerank(request_data_dict, api_key) # Call async method + return self.async_rerank(request_data_dict, api_key, api_base) # Call async method response: Final = client.post( - "https://api.together.xyz/v1/rerank", + _rerank_url(api_base), headers={ "accept": "application/json", "content-type": "application/json", @@ -69,11 +74,12 @@ class TogetherAIRerank(BaseLLM): self, request_data_dict: dict[str, Any], api_key: str, + api_base: str, ) -> RerankResponse: client: Final = get_async_httpx_client(llm_provider=litellm.LlmProviders.TOGETHER_AI) # Use async client response: Final = await client.post( - "https://api.together.xyz/v1/rerank", + _rerank_url(api_base), headers={ "accept": "application/json", "content-type": "application/json", diff --git a/litellm/rerank_api/main.py b/litellm/rerank_api/main.py index 15a6f18a6bb..c8f7842aebf 100644 --- a/litellm/rerank_api/main.py +++ b/litellm/rerank_api/main.py @@ -277,6 +277,8 @@ def rerank( if api_key is None: raise ValueError("TogetherAI API key is required, please set 'TOGETHERAI_API_KEY' in your environment") + api_base = dynamic_api_base or optional_params.api_base or litellm.api_base or "https://api.together.ai/v1" + response = together_rerank.rerank( model=model, query=query, @@ -286,6 +288,7 @@ def rerank( return_documents=return_documents, max_chunks_per_doc=max_chunks_per_doc, api_key=api_key, + api_base=api_base, _is_async=_is_async, ) elif _custom_llm_provider == litellm.LlmProviders.JINA_AI: diff --git a/ruff-strict-budget.json b/ruff-strict-budget.json index 03318718fb5..1ca152985f9 100644 --- a/ruff-strict-budget.json +++ b/ruff-strict-budget.json @@ -1,6 +1,6 @@ { "ANN001": { - "limit": 3018 + "limit": 3016 }, "ANN002": { "limit": 71 @@ -9,7 +9,7 @@ "limit": 827 }, "ANN201": { - "limit": 2016 + "limit": 2015 }, "ANN202": { "limit": 852 @@ -57,7 +57,7 @@ "limit": 3 }, "BLE001": { - "limit": 2919 + "limit": 2918 }, "C401": { "limit": 8 diff --git a/tests/test_litellm/litellm_core_utils/test_get_llm_provider_endpoint_match.py b/tests/test_litellm/litellm_core_utils/test_get_llm_provider_endpoint_match.py index bda7ab4afc6..5c20284282a 100644 --- a/tests/test_litellm/litellm_core_utils/test_get_llm_provider_endpoint_match.py +++ b/tests/test_litellm/litellm_core_utils/test_get_llm_provider_endpoint_match.py @@ -133,3 +133,42 @@ class TestGetLlmProviderRejectsAttackerSmuggledApiBase: assert provider == "groq" assert dynamic_api_key == "server-real-groq-key" + + +class TestTogetherApiBaseResolvesProvider: + """ + Regression for the Together host migration: both the current + ``api.together.ai`` host and the legacy ``api.together.xyz`` host must + resolve to ``together_ai`` when passed as ``api_base``. Before the fix + the endpoint list carried the legacy host but the provider-mapping + chain had no branch for it, so the match fell through with a None + provider and the deployment failed with "LLM Provider NOT provided". + """ + + @pytest.mark.parametrize( + "api_base", + [ + "https://api.together.ai/v1", + "https://api.together.xyz/v1", + ], + ) + def test_together_api_base_resolves_to_together_ai(self, api_base, monkeypatch): + monkeypatch.setenv("TOGETHER_API_KEY", "together-key-from-env") + + model, provider, dynamic_api_key, returned_api_base = get_llm_provider( + model="some-model", + api_base=api_base, + ) + + assert provider == "together_ai" + assert dynamic_api_key == "together-key-from-env" + assert returned_api_base == api_base + assert model == "some-model" + + def test_together_default_api_base_is_together_ai(self, monkeypatch): + monkeypatch.delenv("TOGETHER_AI_API_BASE", raising=False) + + _, provider, _, api_base = get_llm_provider(model="together_ai/some-model") + + assert provider == "together_ai" + assert api_base == "https://api.together.ai/v1" diff --git a/tests/test_litellm/rerank_api/test_main.py b/tests/test_litellm/rerank_api/test_main.py index 85777afe81c..587be59c550 100644 --- a/tests/test_litellm/rerank_api/test_main.py +++ b/tests/test_litellm/rerank_api/test_main.py @@ -1,6 +1,10 @@ import logging from unittest.mock import MagicMock, patch +import httpx +import pytest +import respx + import litellm @@ -62,3 +66,66 @@ def test_rerank_does_not_log_request_content_at_info(caplog): assert all( r.levelno == logging.DEBUG for r in optional_params_logs ), "optional_rerank_params must be logged at DEBUG, not INFO" + + +TOGETHER_RERANK_BODY = { + "id": "rerank-mock-id", + "results": [{"index": 0, "relevance_score": 0.95}], + "usage": {"prompt_tokens": 10, "total_tokens": 10}, +} + + +def test_together_rerank_defaults_to_together_ai_host(respx_mock: respx.MockRouter, monkeypatch): + """Regression for the Together host migration: rerank used to hardcode + https://api.together.xyz/v1/rerank. The default must now be api.together.ai.""" + monkeypatch.delenv("TOGETHER_AI_API_BASE", raising=False) + + mock_route = respx_mock.post("https://api.together.ai/v1/rerank") + mock_route.return_value = httpx.Response(200, json=TOGETHER_RERANK_BODY) + + response = litellm.rerank( + model="together_ai/mixedbread-ai/mxbai-rerank-large-v2", + query=MARKER_QUERY, + documents=[MARKER_DOC], + api_key="fake-together-key", + ) + + assert mock_route.called + assert response.results[0]["relevance_score"] == 0.95 + + +def test_together_rerank_honors_api_base(respx_mock: respx.MockRouter): + """Regression: a custom api_base was silently ignored by the Together rerank handler.""" + mock_route = respx_mock.post("https://custom-together.example/v1/rerank") + mock_route.return_value = httpx.Response(200, json=TOGETHER_RERANK_BODY) + + litellm.rerank( + model="together_ai/mixedbread-ai/mxbai-rerank-large-v2", + query=MARKER_QUERY, + documents=[MARKER_DOC], + api_key="fake-together-key", + api_base="https://custom-together.example/v1", + ) + + assert mock_route.called + assert mock_route.calls[0].request.headers["authorization"] == "Bearer fake-together-key" + + +@pytest.mark.asyncio +async def test_together_rerank_async_honors_env_api_base(respx_mock: respx.MockRouter, monkeypatch): + """Regression: TOGETHER_AI_API_BASE was honored by chat but ignored by rerank.""" + monkeypatch.setenv("TOGETHER_AI_API_BASE", "https://env-together.example/v1") + monkeypatch.setenv("DISABLE_AIOHTTP_TRANSPORT", "True") + + mock_route = respx_mock.post("https://env-together.example/v1/rerank") + mock_route.return_value = httpx.Response(200, json=TOGETHER_RERANK_BODY) + + response = await litellm.arerank( + model="together_ai/mixedbread-ai/mxbai-rerank-large-v2", + query=MARKER_QUERY, + documents=[MARKER_DOC], + api_key="fake-together-key", + ) + + assert mock_route.called + assert response.results[0]["relevance_score"] == 0.95 diff --git a/type-discipline-budget.json b/type-discipline-budget.json index 627811a7f1d..f9fe3042f1e 100644 --- a/type-discipline-budget.json +++ b/type-discipline-budget.json @@ -3,7 +3,7 @@ "limit": 22805 }, "LIT002": { - "limit": 26873 + "limit": 26872 }, "LIT003": { "limit": 269 From 5e6b6c6281f8d26bda113cdd54be0fe71afa4d77 Mon Sep 17 00:00:00 2001 From: mateo-berri <277851410+mateo-berri@users.noreply.github.com> Date: Tue, 25 Aug 2026 10:39:10 -0700 Subject: [PATCH 2/2] fix(together_ai): let an explicit api_key beat the Together env key on api_base match --- litellm/litellm_core_utils/get_llm_provider_logic.py | 2 +- litellm/llms/together_ai/rerank/handler.py | 2 +- .../test_get_llm_provider_endpoint_match.py | 12 ++++++++++++ 3 files changed, 14 insertions(+), 2 deletions(-) diff --git a/litellm/litellm_core_utils/get_llm_provider_logic.py b/litellm/litellm_core_utils/get_llm_provider_logic.py index d2d82064c47..005e94ebe82 100644 --- a/litellm/litellm_core_utils/get_llm_provider_logic.py +++ b/litellm/litellm_core_utils/get_llm_provider_logic.py @@ -274,7 +274,7 @@ def get_llm_provider( dynamic_api_key = get_secret_str("DEEPSEEK_API_KEY") elif endpoint == "api.together.ai/v1" or endpoint == "api.together.xyz/v1": custom_llm_provider = "together_ai" - dynamic_api_key = ( + dynamic_api_key = api_key or ( get_secret_str("TOGETHER_API_KEY") or get_secret_str("TOGETHER_AI_API_KEY") or get_secret_str("TOGETHERAI_API_KEY") diff --git a/litellm/llms/together_ai/rerank/handler.py b/litellm/llms/together_ai/rerank/handler.py index 8407018b898..b8079e52c97 100644 --- a/litellm/llms/together_ai/rerank/handler.py +++ b/litellm/llms/together_ai/rerank/handler.py @@ -51,7 +51,7 @@ class TogetherAIRerank(BaseLLM): raise ValueError("TogetherAI does not support max_chunks_per_doc") if _is_async: - return self.async_rerank(request_data_dict, api_key, api_base) # Call async method + return self.async_rerank(request_data_dict, api_key, api_base) response: Final = client.post( _rerank_url(api_base), diff --git a/tests/test_litellm/litellm_core_utils/test_get_llm_provider_endpoint_match.py b/tests/test_litellm/litellm_core_utils/test_get_llm_provider_endpoint_match.py index 5c20284282a..6cacd119030 100644 --- a/tests/test_litellm/litellm_core_utils/test_get_llm_provider_endpoint_match.py +++ b/tests/test_litellm/litellm_core_utils/test_get_llm_provider_endpoint_match.py @@ -165,6 +165,18 @@ class TestTogetherApiBaseResolvesProvider: assert returned_api_base == api_base assert model == "some-model" + def test_explicit_api_key_beats_together_env_key(self, monkeypatch): + monkeypatch.setenv("TOGETHER_API_KEY", "together-key-from-env") + + _, provider, dynamic_api_key, _ = get_llm_provider( + model="some-model", + api_base="https://api.together.ai/v1", + api_key="explicit-caller-key", + ) + + assert provider == "together_ai" + assert dynamic_api_key == "explicit-caller-key" + def test_together_default_api_base_is_together_ai(self, monkeypatch): monkeypatch.delenv("TOGETHER_AI_API_BASE", raising=False)