qi[&dZddlmZddlmZddlmZddlmZddl m Z ddl m Z m Z ddlmZdd lmZdd lmZmZmZdd lmZeGd d ZeGddeZGddeZy)zGrok LLM service implementation using OpenAI-compatible interface. This module provides a service for interacting with Grok's API through an OpenAI-compatible interface, including specialized token usage tracking and context aggregation functionality. ) dataclass)Optional)logger) LLMTokenUsage) LLMContext)LLMAssistantAggregatorParamsLLMUserAggregatorParams)OpenAILLMContext)OpenAILLMSettings) OpenAIAssistantContextAggregatorOpenAILLMServiceOpenAIUserContextAggregator)_warn_deprecated_paramc>eZdZUdZeed<eed<defdZdefdZy)GrokContextAggregatorPairaTPair of context aggregators for user and assistant interactions. Provides a convenient container for managing both user and assistant context aggregators together for Grok LLM interactions. .. deprecated:: 0.0.99 `GrokContextAggregatorPair` is deprecated and will be removed in a future version. Use the universal `LLMContext` and `LLMContextAggregatorPair` instead. See `OpenAILLMContext` docstring for migration guide. Parameters: _user: The user context aggregator instance. _assistant: The assistant context aggregator instance. _user _assistantreturnc|jS)zmGet the user context aggregator. Returns: The user context aggregator instance. )rselfs K/opt/pipecat/venv/lib/python3.12/site-packages/pipecat/services/grok/llm.pyuserzGrokContextAggregatorPair.user8s zzc|jS)zwGet the assistant context aggregator. Returns: The assistant context aggregator instance. )rrs r assistantz#GrokContextAggregatorPair.assistant@s rN) __name__ __module__ __qualname____doc__r__annotations__r rrrrrr#s/  '&001;rrceZdZdZy)GrokLLMSettingszSettings for GrokLLMService.N)rrrr r"rrr$r$Is&rr$c eZdZUdZeZeed<dddddededeed eeffd Z dfd Z d e e zffd Z defdZeedd e dededefdZxZS)GrokLLMServiceamA service for interacting with Grok's API using the OpenAI-compatible interface. This service extends OpenAILLMService to connect to Grok's API endpoint while maintaining full compatibility with OpenAI's interface and functionality. Includes specialized token usage tracking that accumulates metrics during processing and reports final totals. _settingszhttps://api.x.ai/v1N)base_urlmodelsettingsapi_keyr(r)r*c td}|tdtd||_||j|t |d|||d|d|_d|_d|_d|_ d|_ y) aInitialize the GrokLLMService with API key and model. Args: api_key: The API key for accessing Grok's API. base_url: The base URL for Grok API. Defaults to "https://api.x.ai/v1". model: The model identifier to use. Defaults to "grok-3-beta". .. deprecated:: 0.0.105 Use ``settings=OpenAILLMSettings(model=...)`` instead. settings: Runtime-updatable settings. When provided alongside deprecated parameters, ``settings`` values take precedence. **kwargs: Additional keyword arguments passed to OpenAILLMService. z grok-3-beta)r)Nr))r+r(r*rFr") r$rr) apply_updatesuper__init___prompt_tokens_completion_tokens _total_tokens_has_reported_prompt_tokens_is_processing)rr+r(r)r*kwargsdefault_settings __class__s rr/zGrokLLMService.__init__\s0+?   "7OW E%*  "    ) )( 3 a8FVaZ`a"#+0(#rc Vtjd|t| ||fi|S)asCreate OpenAI-compatible client for Grok API endpoint. Args: api_key: The API key to use. If None, uses instance default. base_url: The base URL to use. If None, uses instance default. **kwargs: Additional arguments passed to client creation. Returns: The configured client instance for Grok API. zCreating Grok client with api )rdebugr. create_client)rr+r(r5r7s rr:zGrokLLMService.create_clients0  5hZ@Aw$WhA&AArcontextc (Kd|_d|_d|_d|_d|_d|_d|_ t|!|d{d|_|jdkDs|jdkDrx|j|jz|_t|j|j|j|j|j}t|)|d{yy77#d|_|jdkDs|jdkDry|j|jz|_t|j|j|j|j|j}t|)|d{7wwxYww)aProcess a context through the LLM and accumulate token usage metrics. This method overrides the parent class implementation to handle Grok's incremental token reporting style, accumulating the counts and reporting them once at the end of processing. Args: context: The context to process, containing messages and other information needed for the LLM interaction. rNFT) prompt_tokenscompletion_tokens total_tokenscache_read_input_tokensreasoning_tokens) r0r1r2_cache_read_input_tokens_reasoning_tokensr3r4r._process_contextrstart_llm_usage_metrics)rr;tokensr7s rrDzGrokLLMService._process_contexts "#(,%!%+0(" >'*73 3 3"'D ""Q&$*A*AA*E%)%8%84;R;R%R"&"&"5"5&*&=&=!%!3!3,0,I,I%)%;%; g5f===+F 4>#(D ""Q&$*A*AA*E%)%8%84;R;R%R"&"&"5"5&*&=&=!%!3!3,0,I,I%)%;%; g5f===+FsN2FC/C+ C/ BF$C-%F+C/-F/BFF FFrFcVK|jsy|js'|jdkDr|j|_d|_|j|j kDr|j|_|j |j |_|j|j|_ yyw)aAccumulate token usage metrics during processing. This method intercepts the incremental token updates from Grok's API and accumulates them instead of passing each update to the metrics system. The final accumulated totals are reported at the end of processing. Args: tokens: The token usage metrics for the current chunk of processing, containing prompt_tokens, completion_tokens, and optional cached/reasoning tokens. NrT) r4r3r=r0r>r1r@rBrArC)rrFs rrEz&GrokLLMService.start_llm_usage_metricss"" //F4H4H14L"("6"6D /3D ,  # #d&=&= =&,&>&>D #  ) ) 5,2,J,JD )  " " .%+%<%.>.K">H=M=F0G/H9U9W !K!!K- !K 7 !K # !Krr&N)r dataclassesrtypingrlogururpipecat.metrics.metricsr*pipecat.processors.aggregators.llm_contextr+pipecat.processors.aggregators.llm_responserr 1pipecat.processors.aggregators.openai_llm_contextr pipecat.services.openai.base_llmr pipecat.services.openai.llmr r rpipecat.services.settingsrrr$r&r"rrr\s"1AO> = "" "J  '   kK%kKr