README.md

FriendliCoreServerlessToken

(serverless.token)

Overview

Available Operations

• tokenization - Tokenization
• detokenization - Detokenization

tokenization

By giving a text input, generate a tokenized output of token IDs.

Example Usage

import os

from friendli import SyncFriendli

with SyncFriendli(
    token=os.getenv("FRIENDLI_TOKEN", ""),
) as friendli:
    res = friendli.serverless.token.tokenization(
        model="meta-llama-3.1-8b-instruct", prompt="What is generative AI?"
    )

    # Handle response
    print(res)

Parameters

Parameter	Type	Required	Description	Example
model	str	✔️	Code of the model to use. See available model list.	meta-llama-3.1-8b-instruct
prompt	str	✔️	Input text prompt to tokenize.	What is generative AI?
x_friendli_team	OptionalNullable[str]	➖	ID of team to run requests as (optional parameter).
retries	Optional[utils.RetryConfig]	➖	Configuration to override the default retry behavior of the client.

Response

models.ServerlessTokenizationSuccess

Errors

Error Type	Status Code	Content Type
models.SDKError	4XX, 5XX	*/*

detokenization

By giving a list of tokens, generate a detokenized output text string.

Example Usage

import os

from friendli import SyncFriendli

with SyncFriendli(
    token=os.getenv("FRIENDLI_TOKEN", ""),
) as friendli:
    res = friendli.serverless.token.detokenization(
        model="meta-llama-3.1-8b-instruct",
        tokens=[
            128000,
            3923,
            374,
            1803,
            1413,
            15592,
            30,
        ],
    )

    # Handle response
    print(res)

Parameters

Parameter	Type	Required	Description	Example
model	str	✔️	Code of the model to use. See available model list.	meta-llama-3.1-8b-instruct
tokens	List[int]	✔️	A token sequence to detokenize.	[ 128000, 3923, 374, 1803, 1413, 15592, 30 ]
x_friendli_team	OptionalNullable[str]	➖	ID of team to run requests as (optional parameter).
retries	Optional[utils.RetryConfig]	➖	Configuration to override the default retry behavior of the client.

Response

models.ServerlessDetokenizationSuccess

Errors

Error Type	Status Code	Content Type
models.SDKError	4XX, 5XX	*/*