From d930a46e648e6e78a08b4fe5828e3f6796c09b6b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Thu, 24 Oct 2024 12:57:46 -0700
Subject: [PATCH] services(together): fix together AI InputParams

---
 src/pipecat/services/openai.py   |  3 +++
 src/pipecat/services/together.py | 41 ++------------------------------
 2 files changed, 5 insertions(+), 39 deletions(-)

diff --git a/src/pipecat/services/openai.py b/src/pipecat/services/openai.py
index e12744fcd..598240bd5 100644
--- a/src/pipecat/services/openai.py
+++ b/src/pipecat/services/openai.py
@@ -99,6 +99,9 @@ class BaseOpenAILLMService(LLMService):
         )
         seed: Optional[int] = Field(default_factory=lambda: NOT_GIVEN, ge=0)
         temperature: Optional[float] = Field(default_factory=lambda: NOT_GIVEN, ge=0.0, le=2.0)
+        # Note: top_k is currently not supported by the OpenAI client library,
+        # so top_k is ignore right now.
+        top_k: Optional[int] = Field(default=None, ge=0)
         top_p: Optional[float] = Field(default_factory=lambda: NOT_GIVEN, ge=0.0, le=1.0)
         max_tokens: Optional[int] = Field(default_factory=lambda: NOT_GIVEN, ge=1)
         max_completion_tokens: Optional[int] = Field(default_factory=lambda: NOT_GIVEN, ge=1)
diff --git a/src/pipecat/services/together.py b/src/pipecat/services/together.py
index 1bf74e508..e2aed4da1 100644
--- a/src/pipecat/services/together.py
+++ b/src/pipecat/services/together.py
@@ -4,11 +4,8 @@
 # SPDX-License-Identifier: BSD 2-Clause License
 #
 
-from typing import Any, Dict, Optional
 
-import httpx
 from loguru import logger
-from pydantic import BaseModel, Field
 
 from pipecat.services.openai import OpenAILLMService
 
@@ -27,50 +24,16 @@ except ModuleNotFoundError as e:
 class TogetherLLMService(OpenAILLMService):
     """This class implements inference with Together's Llama 3.1 models"""
 
-    class InputParams(BaseModel):
-        frequency_penalty: Optional[float] = Field(default=None, ge=-2.0, le=2.0)
-        max_tokens: Optional[int] = Field(default=4096, ge=1)
-        presence_penalty: Optional[float] = Field(default=None, ge=-2.0, le=2.0)
-        temperature: Optional[float] = Field(default=None, ge=0.0, le=1.0)
-        # Note: top_k is currently not supported by the OpenAI client library,
-        # so top_k is ignore right now.
-        top_k: Optional[int] = Field(default=None, ge=0)
-        top_p: Optional[float] = Field(default=None, ge=0.0, le=1.0)
-        extra: Optional[Dict[str, Any]] = Field(default_factory=dict)
-        seed: Optional[int] = Field(default=None)
-
     def __init__(
         self,
         *,
         api_key: str,
         base_url: str = "https://api.together.xyz/v1",
         model: str = "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo",
-        params: InputParams = InputParams(),
         **kwargs,
     ):
-        super().__init__(api_key=api_key, base_url=base_url, model=model, params=params, **kwargs)
-        self.set_model_name(model)
-        self._settings = {
-            "max_tokens": params.max_tokens,
-            "frequency_penalty": params.frequency_penalty,
-            "presence_penalty": params.presence_penalty,
-            "seed": params.seed,
-            "temperature": params.temperature,
-            "top_p": params.top_p,
-            "extra": params.extra if isinstance(params.extra, dict) else {},
-        }
-
-    def can_generate_metrics(self) -> bool:
-        return True
+        super().__init__(api_key=api_key, base_url=base_url, model=model, **kwargs)
 
     def create_client(self, api_key=None, base_url=None, **kwargs):
         logger.debug(f"Creating Together.ai client with api {base_url}")
-        return AsyncOpenAI(
-            api_key=api_key,
-            base_url=base_url,
-            http_client=DefaultAsyncHttpxClient(
-                limits=httpx.Limits(
-                    max_keepalive_connections=100, max_connections=1000, keepalive_expiry=None
-                )
-            ),
-        )
+        return super().create_client(api_key, base_url, **kwargs)