Vllm.Outputs.EmbeddingRequestOutput (VLLM v0.3.0)

Copy Markdown View Source

The output data of a pooling request to the LLM.

Parameters

  • request_id - A unique identifier for the pooling request. (type: String.t())
  • outputs - The pooling results for the given input. (type: PoolingOutput)
  • prompt_token_ids - A list of token IDs used in the prompt. (type: list(integer()))
  • num_cached_tokens - The number of tokens with prefix cache hit.
  • finished - A flag indicating whether the pooling is completed. (type: boolean())

Summary

Functions

Python method EmbeddingRequestOutput.from_base.

Initialize self. See help(type(self)) for accurate signature.

Types

t()

@opaque t()

Functions

from_base(ref, request_output, opts \\ [])

@spec from_base(SnakeBridge.Ref.t(), Vllm.Outputs.PoolingRequestOutput.t(), keyword()) ::
  {:ok, term()} | {:error, Snakepit.Error.t()}

Python method EmbeddingRequestOutput.from_base.

Parameters

  • request_output (Vllm.Outputs.PoolingRequestOutput.t())

Returns

  • term()

new(request_id, outputs, prompt_token_ids, num_cached_tokens, finished, opts \\ [])

@spec new(String.t(), term(), [integer()], integer(), boolean(), keyword()) ::
  {:ok, SnakeBridge.Ref.t()} | {:error, Snakepit.Error.t()}

Initialize self. See help(type(self)) for accurate signature.

Parameters

  • request_id (String.t())
  • outputs (term())
  • prompt_token_ids (list(integer()))
  • num_cached_tokens (integer())
  • finished (boolean())