alignment with pr 4081

README.md

@@ -84,7 +84,7 @@ Catch new features, interviews, and how-tos on our [Pipecat TV](https://www.yout
 | Category            | Services                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   |
 | ------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
 | Speech-to-Text      | [AssemblyAI](https://docs.pipecat.ai/server/services/stt/assemblyai), [AWS](https://docs.pipecat.ai/server/services/stt/aws), [Azure](https://docs.pipecat.ai/server/services/stt/azure), [Cartesia](https://docs.pipecat.ai/server/services/stt/cartesia), [Deepgram](https://docs.pipecat.ai/server/services/stt/deepgram), [ElevenLabs](https://docs.pipecat.ai/server/services/stt/elevenlabs), [Fal Wizper](https://docs.pipecat.ai/server/services/stt/fal), [Gladia](https://docs.pipecat.ai/server/services/stt/gladia), [Google](https://docs.pipecat.ai/server/services/stt/google), [Gradium](https://docs.pipecat.ai/server/services/stt/gradium), [Groq (Whisper)](https://docs.pipecat.ai/server/services/stt/groq), [NVIDIA Riva](https://docs.pipecat.ai/server/services/stt/riva), [OpenAI (Whisper)](https://docs.pipecat.ai/server/services/stt/openai), [SambaNova (Whisper)](https://docs.pipecat.ai/server/services/stt/sambanova), [Sarvam](https://docs.pipecat.ai/server/services/stt/sarvam), [Soniox](https://docs.pipecat.ai/server/services/stt/soniox), [Speechmatics](https://docs.pipecat.ai/server/services/stt/speechmatics), [Whisper](https://docs.pipecat.ai/server/services/stt/whisper)                                                                                                                                                                                                                                                                                                                             |
-| LLMs                | [Anthropic](https://docs.pipecat.ai/server/services/llm/anthropic), [AWS](https://docs.pipecat.ai/server/services/llm/aws), [Azure](https://docs.pipecat.ai/server/services/llm/azure), [Cerebras](https://docs.pipecat.ai/server/services/llm/cerebras), [DeepSeek](https://docs.pipecat.ai/server/services/llm/deepseek), [Fireworks AI](https://docs.pipecat.ai/server/services/llm/fireworks), [Gemini](https://docs.pipecat.ai/server/services/llm/gemini), [Grok](https://docs.pipecat.ai/server/services/llm/grok), [Groq](https://docs.pipecat.ai/server/services/llm/groq), [Mistral](https://docs.pipecat.ai/server/services/llm/mistral), [NVIDIA NIM](https://docs.pipecat.ai/server/services/llm/nim), [Ollama](https://docs.pipecat.ai/server/services/llm/ollama), [OpenAI](https://docs.pipecat.ai/server/services/llm/openai), [OpenRouter](https://docs.pipecat.ai/server/services/llm/openrouter), [Perplexity](https://docs.pipecat.ai/server/services/llm/perplexity), [Qwen](https://docs.pipecat.ai/server/services/llm/qwen), [SambaNova](https://docs.pipecat.ai/server/services/llm/sambanova) [Together AI](https://docs.pipecat.ai/server/services/llm/together)                                                                                                                                                                                                                                                                                                                                                               |
+| LLMs                | [Anthropic](https://docs.pipecat.ai/server/services/llm/anthropic), [AWS](https://docs.pipecat.ai/server/services/llm/aws), [Azure](https://docs.pipecat.ai/server/services/llm/azure), [Cerebras](https://docs.pipecat.ai/server/services/llm/cerebras), [DeepSeek](https://docs.pipecat.ai/server/services/llm/deepseek), [Fireworks AI](https://docs.pipecat.ai/server/services/llm/fireworks), [Gemini](https://docs.pipecat.ai/server/services/llm/gemini), [Grok](https://docs.pipecat.ai/server/services/llm/grok), [Groq](https://docs.pipecat.ai/server/services/llm/groq), [Mistral](https://docs.pipecat.ai/server/services/llm/mistral), [NVIDIA NIM](https://docs.pipecat.ai/server/services/llm/nim), [Ollama](https://docs.pipecat.ai/server/services/llm/ollama), [OpenAI](https://docs.pipecat.ai/server/services/llm/openai), [OpenRouter](https://docs.pipecat.ai/server/services/llm/openrouter), [Perplexity](https://docs.pipecat.ai/server/services/llm/perplexity), [Qwen](https://docs.pipecat.ai/server/services/llm/qwen), [SambaNova](https://docs.pipecat.ai/server/services/llm/sambanova), [Sarvam](https://docs.pipecat.ai/server/services/llm/sarvam), [Together AI](https://docs.pipecat.ai/server/services/llm/together)                                                                                                                                                                                                                                                                                                                                                               |
 | Text-to-Speech      | [Async](https://docs.pipecat.ai/server/services/tts/asyncai), [AWS](https://docs.pipecat.ai/server/services/tts/aws), [Azure](https://docs.pipecat.ai/server/services/tts/azure), [Camb AI](https://docs.pipecat.ai/server/services/tts/camb), [Cartesia](https://docs.pipecat.ai/server/services/tts/cartesia), [Deepgram](https://docs.pipecat.ai/server/services/tts/deepgram), [ElevenLabs](https://docs.pipecat.ai/server/services/tts/elevenlabs), [Fish](https://docs.pipecat.ai/server/services/tts/fish), [Google](https://docs.pipecat.ai/server/services/tts/google), [Gradium](https://docs.pipecat.ai/server/services/tts/gradium), [Groq](https://docs.pipecat.ai/server/services/tts/groq), [Hume](https://docs.pipecat.ai/server/services/tts/hume), [Inworld](https://docs.pipecat.ai/server/services/tts/inworld), [LMNT](https://docs.pipecat.ai/server/services/tts/lmnt), [MiniMax](https://docs.pipecat.ai/server/services/tts/minimax), [Neuphonic](https://docs.pipecat.ai/server/services/tts/neuphonic), [NVIDIA Riva](https://docs.pipecat.ai/server/services/tts/riva), [OpenAI](https://docs.pipecat.ai/server/services/tts/openai), [Piper](https://docs.pipecat.ai/server/services/tts/piper), [Resemble](https://docs.pipecat.ai/server/services/tts/resemble), [Rime](https://docs.pipecat.ai/server/services/tts/rime), [Sarvam](https://docs.pipecat.ai/server/services/tts/sarvam), [Speechmatics](https://docs.pipecat.ai/server/services/tts/speechmatics), [XTTS](https://docs.pipecat.ai/server/services/tts/xtts) |
 | Speech-to-Speech    | [AWS Nova Sonic](https://docs.pipecat.ai/server/services/s2s/aws), [Gemini Multimodal Live](https://docs.pipecat.ai/server/services/s2s/gemini), [Grok Voice Agent](https://docs.pipecat.ai/server/services/s2s/grok), [OpenAI Realtime](https://docs.pipecat.ai/server/services/s2s/openai), [Ultravox](https://docs.pipecat.ai/server/services/s2s/ultravox),                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                            |
 | Transport           | [Daily (WebRTC)](https://docs.pipecat.ai/server/services/transport/daily), [FastAPI Websocket](https://docs.pipecat.ai/server/services/transport/fastapi-websocket), [SmallWebRTCTransport](https://docs.pipecat.ai/server/services/transport/small-webrtc), [WebSocket Server](https://docs.pipecat.ai/server/services/transport/websocket-server), Local                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                 |

examples/foundational/07z-interruptible-sarvam-http.py

@@ -23,7 +23,7 @@ from pipecat.processors.aggregators.llm_response_universal import (
 )
 from pipecat.runner.types import RunnerArguments
 from pipecat.runner.utils import create_transport
-from pipecat.services.openai.llm import OpenAILLMService
+from pipecat.services.sarvam.llm import SarvamLLMService
 from pipecat.services.sarvam.stt import SarvamSTTService
 from pipecat.services.sarvam.tts import SarvamHttpTTSService
 from pipecat.transcriptions.language import Language
@@ -72,10 +72,10 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
             ),
         )
 
-        llm = OpenAILLMService(
+        llm = SarvamLLMService(
-            api_key=os.getenv("OPENAI_API_KEY"),
+            api_key=os.getenv("SARVAM_API_KEY"),
-            settings=OpenAILLMService.Settings(
+            settings=SarvamLLMService.Settings(
-                system_instruction="You are a helpful LLM in a WebRTC call. Your goal is to demonstrate your capabilities in a succinct way. Your output will be spoken aloud, so avoid special characters that can't easily be spoken, such as emojis or bullet points. Respond to what the user said in a creative and helpful way.",
+                system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
             ),
         )
 

examples/foundational/07z-interruptible-sarvam.py

@@ -13,7 +13,7 @@ from pipecat.audio.vad.silero import SileroVADAnalyzer
 from pipecat.frames.frames import LLMRunFrame
 from pipecat.pipeline.pipeline import Pipeline
 from pipecat.pipeline.runner import PipelineRunner
-from pipecat.pipeline.task import PipelineParams, PipelineTask
+from pipecat.pipeline.task import IDLE_TIMEOUT_SECS, PipelineParams, PipelineTask
 from pipecat.processors.aggregators.llm_context import LLMContext
 from pipecat.processors.aggregators.llm_response_universal import (
     LLMContextAggregatorPair,
@@ -21,7 +21,7 @@ from pipecat.processors.aggregators.llm_response_universal import (
 )
 from pipecat.runner.types import RunnerArguments
 from pipecat.runner.utils import create_transport
-from pipecat.services.openai.llm import OpenAILLMService
+from pipecat.services.sarvam.llm import SarvamLLMService
 from pipecat.services.sarvam.stt import SarvamSTTService
 from pipecat.services.sarvam.tts import SarvamTTSService
 from pipecat.transports.base_transport import BaseTransport, TransportParams
@@ -55,21 +55,21 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     stt = SarvamSTTService(
         api_key=os.getenv("SARVAM_API_KEY"),
         settings=SarvamSTTService.Settings(
-            model="saarika:v2.5",
+            model="saaras:v3",
         ),
     )
 
     tts = SarvamTTSService(
         api_key=os.getenv("SARVAM_API_KEY"),
         settings=SarvamTTSService.Settings(
-            model="bulbul:v2",
+            model="bulbul:v3",
-            voice="manisha",
+            voice="shubh",
         ),
     )
-    llm = OpenAILLMService(
+    llm = SarvamLLMService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.getenv("SARVAM_API_KEY"),
-        settings=OpenAILLMService.Settings(
+        settings=SarvamLLMService.Settings(
-            system_instruction="You are a helpful LLM in a WebRTC call. Your goal is to demonstrate your capabilities in a succinct way. Your output will be spoken aloud, so avoid special characters that can't easily be spoken, such as emojis or bullet points. Respond to what the user said in a creative and helpful way.",
+            model="sarvam-30b",
         ),
     )
 
@@ -96,6 +96,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
         params=PipelineParams(
             enable_metrics=True,
             enable_usage_metrics=True,
+            allow_interruptions=True,
         ),
     )
 

scripts/evals/run-release-evals.py

@@ -184,6 +184,7 @@ TESTS_14 = [
     ("14v-function-calling-openai.py", EVAL_WEATHER),
     ("14w-function-calling-mistral.py", EVAL_WEATHER),
     ("14x-function-calling-openpipe.py", EVAL_WEATHER),
+    ("14y-function-calling-sarvam.py", EVAL_WEATHER),
     # Video
     ("14d-function-calling-anthropic-video.py", EVAL_VISION_CAMERA),
     ("14d-function-calling-aws-video.py", EVAL_VISION_CAMERA),

src/pipecat/services/sarvam/llm.py

@@ -36,19 +36,16 @@ class SarvamLLMSettings(OpenAILLMSettings):
 
 
 class SarvamLLMService(OpenAILLMService):
-    """Sarvam LLM service using Sarvam's OpenAI-compatible chat completions API.
+    """A service for interacting with Sarvam's API using the OpenAI-compatible interface.
 
-    This service extends ``OpenAILLMService`` while adding Sarvam-specific behavior:
+    This service extends OpenAILLMService to connect to Sarvam's API endpoint while
-
+    maintaining full compatibility with OpenAI's interface and functionality.
-    - model allow-list validation
-    - request shaping for Sarvam-compatible parameters
-    - Sarvam auth header wiring (``api-subscription-key``)
-    - SDK User-Agent propagation on every API call
     """
 
     _SUPPORTED_MODELS = frozenset(
         {"sarvam-30b", "sarvam-30b-16k", "sarvam-105b", "sarvam-105b-32k"}
     )
+    _TOOL_CALLING_MODELS = _SUPPORTED_MODELS
     Settings = SarvamLLMSettings
     _settings: Settings
 
@@ -70,29 +67,20 @@ class SarvamLLMService(OpenAILLMService):
             default_headers: Additional HTTP headers to include in requests.
             **kwargs: Additional keyword arguments passed to ``OpenAILLMService``.
         """
-        # Initialize defaults with concrete values for Sarvam-specific fields.
+        # Initialize only Sarvam-specific defaults; inherited defaults are
+        # provided by the OpenAI base service initialization.
         default_settings = self.Settings(
             model="sarvam-30b",
-            system_instruction=None,
-            frequency_penalty=NOT_GIVEN,
-            presence_penalty=NOT_GIVEN,
-            seed=NOT_GIVEN,
-            temperature=NOT_GIVEN,
-            top_p=NOT_GIVEN,
-            top_k=None,
-            max_tokens=NOT_GIVEN,
-            max_completion_tokens=NOT_GIVEN,
-            filter_incomplete_user_turns=False,
-            user_turn_completion_config=None,
-            extra={},
             wiki_grounding=None,
             reasoning_effort=None,
         )
 
-        # Apply settings delta (canonical API, always wins).
+        # Apply settings delta (canonical API, always wins)
         if settings is not None:
             default_settings.apply_update(settings)
 
+        self._validate_model(default_settings.model)
+
         super().__init__(
             api_key=api_key,
             base_url=base_url,
@@ -100,9 +88,12 @@ class SarvamLLMService(OpenAILLMService):
             default_headers=default_headers,
             **kwargs,
         )
-        # Keep Sarvam-specific settings object so runtime updates include
+        # Rehydrate into Sarvam settings using inherited concrete store values
-        # ``wiki_grounding`` and ``reasoning_effort`` without extra bridging.
+        # from base initialization, then layer Sarvam-specific fields.
-        self._settings = default_settings
+        sarvam_settings = self.Settings.from_mapping(self._settings.given_fields())
+        sarvam_settings.wiki_grounding = default_settings.wiki_grounding
+        sarvam_settings.reasoning_effort = default_settings.reasoning_effort
+        self._settings = sarvam_settings
 
     def create_client(
         self,
@@ -156,43 +147,10 @@ class SarvamLLMService(OpenAILLMService):
 
         return params
 
-    async def _call_with_raw_sarvam_errors(self, awaitable: Awaitable[_T]) -> _T:
+    def _validate_model(self, model: str):
-        """Await an OpenAI call while preserving Sarvam raw error payloads.
+        if model not in self._SUPPORTED_MODELS:
-
+            allowed = ", ".join(sorted(self._SUPPORTED_MODELS))
-        BaseOpenAILLMService handles pipeline-frame exceptions via push_error(),
+            raise ValueError(f"Unsupported Sarvam LLM model '{model}'. Allowed values: {allowed}.")
-        but direct helper methods like ``get_chat_completions`` and
-        ``run_inference`` are often consumed directly. We normalize those errors
-        here so applications consistently receive server-provided messages.
-        """
-        try:
-            return await awaitable
-        except (APITimeoutError, asyncio.TimeoutError, httpx.TimeoutException):
-            raise
-        except Exception as e:
-            raise RuntimeError(str(e)) from e
-
-    async def get_chat_completions(
-        self, params_from_context: OpenAILLMInvocationParams
-    ) -> AsyncStream[ChatCompletionChunk]:
-        """Get streaming chat completions with Sarvam raw error passthrough."""
-        return await self._call_with_raw_sarvam_errors(
-            super().get_chat_completions(params_from_context)
-        )
-
-    async def run_inference(
-        self,
-        context: LLMContext | OpenAILLMContext,
-        max_tokens: Optional[int] = None,
-        system_instruction: Optional[str] = None,
-    ) -> Optional[str]:
-        """Run one-shot inference and preserve Sarvam raw server errors."""
-        return await self._call_with_raw_sarvam_errors(
-            super().run_inference(
-                context,
-                max_tokens=max_tokens,
-                system_instruction=system_instruction,
-            )
-        )
 
     def _validate_tool_parameters(self, params_from_context: OpenAILLMInvocationParams):
         tools = params_from_context.get("tools", NOT_GIVEN)