Detokenize

Generates text from tokens using a model's tokenizer.

This endpoint expects an object.

modelstringRequired

The name of the model to use for detokenization.

tokenslist of integersRequired

The tokens to turn into text.

Successful response.

idstring or null

Unique ID for the token.

objectstring or null

Type of object (tokens).

createdinteger or null

Timestamp of when the tokens were generated.

modelstring or null

The name of the model used.

textstring or null

The converted tokens

1	curl -X POST https://api.predictionguard.com/detokenize \
2	-H "Authorization: Bearer <token>" \
3	-H "Content-Type: application/json" \
4	-d '{
5	"model": "Hermes-3-Llama-3.1-70B",
6	"tokens": [
7	1,
8	851,
9	349,
10	264,
11	6029,
12	653,
13	2757,
14	28723
15	]
16	}'

1	{
2	"id": "token-ab046fcf-945f-421c-b9f0-1c75ff355203",
3	"object": "tokens",
4	"created": 1729871708,
5	"model": "Hermes-3-Llama-3.1-70B",
6	"text": "This is a tokenize example."
7	}