More completions api update (#148)

* update client * lint

Author: yunfeng-scale

launch/client.py

@@ -2898,6 +2898,8 @@ def completions_sync(
  prompt: str,
  max_new_tokens: int,
  temperature: float,
+ stop_sequences: Optional[List[str]] = None,
+ return_token_log_probs: Optional[bool] = False,
  ) -> CompletionSyncV1Response:
  """
  Run prompt completion on a sync LLM endpoint. Will fail if the endpoint is not sync.
@@ -2911,12 +2913,22 @@ def completions_sync(
 
  temperature: The temperature to use for sampling
 
+ stop_sequences: List of sequences to stop the completion at
+
+ return_token_log_probs: Whether to return the log probabilities of the tokens
+
  Returns:
  Response for prompt completion
  """
  with ApiClient(self.configuration) as api_client:
  api_instance = DefaultApi(api_client)
- request = CompletionSyncV1Request(max_new_tokens=max_new_tokens, prompt=prompt, temperature=temperature)
+ request = CompletionSyncV1Request(
+ max_new_tokens=max_new_tokens,
+ prompt=prompt,
+ temperature=temperature,
+ stop_sequences=stop_sequences if stop_sequences is not None else [],
+ return_token_log_probs=return_token_log_probs,
+ )
  query_params = frozendict({"model_endpoint_name": endpoint_name})
  response = api_instance.create_completion_sync_task_v1_llm_completions_sync_post( # type: ignore
  body=request,
@@ -2932,6 +2944,8 @@ def completions_stream(
  prompt: str,
  max_new_tokens: int,
  temperature: float,
+ stop_sequences: Optional[List[str]] = None,
+ return_token_log_probs: Optional[bool] = False,
  ) -> Iterable[CompletionStreamV1Response]:
  """
  Run prompt completion on an LLM endpoint in streaming fashion. Will fail if endpoint does not support streaming.
@@ -2945,10 +2959,20 @@ def completions_stream(
 
  temperature: The temperature to use for sampling
 
+ stop_sequences: List of sequences to stop the completion at
+
+ return_token_log_probs: Whether to return the log probabilities of the tokens
+
  Returns:
  Iterable responses for prompt completion
  """
- request = {"max_new_tokens": max_new_tokens, "prompt": prompt, "temperature": temperature}
+ request = {
+ "max_new_tokens": max_new_tokens,
+ "prompt": prompt,
+ "temperature": temperature,
+ "stop_sequences": stop_sequences,
+ "return_token_log_probs": return_token_log_probs,
+ }
  response = requests.post(
  url=f"{self.configuration.host}/v1/llm/completions-stream?model_endpoint_name={endpoint_name}",
  json=request,