huggingface.js/pr_2407/en/inference/interfaces/TextGenerationStreamDetails.md

Interface: TextGenerationStreamDetails

Properties

best_of_sequences

• Optional best_of_sequences: TextGenerationStreamBestOfSequence[]

Additional sequences when using the best_of parameter

Defined in[[bestofsequences.defined-in]]

inference/src/tasks/nlp/textGenerationStream.ts:68

---

finish_reason

• finish_reason: TextGenerationStreamFinishReason

Generation finish reason

Defined in[[finishreason.defined-in]]

inference/src/tasks/nlp/textGenerationStream.ts:58

---

generated_tokens

• generated_tokens: number

Number of generated tokens

Defined in[[generatedtokens.defined-in]]

inference/src/tasks/nlp/textGenerationStream.ts:60

---

prefill

• prefill: TextGenerationStreamPrefillToken[]

Prompt tokens

Defined in[[prefill.defined-in]]

inference/src/tasks/nlp/textGenerationStream.ts:64

---

seed

• Optional seed: number

Sampling seed if sampling was activated

Defined in[[seed.defined-in]]

inference/src/tasks/nlp/textGenerationStream.ts:62

---

tokens

• tokens: TextGenerationStreamToken[]

Defined in[[tokens.defined-in]]

inference/src/tasks/nlp/textGenerationStream.ts:66