GET

/v1/schemas/fireworks/chat/v1/completions?kind=output

Accept:

kind outputactivity chathash 5d9d4233…extracted Sep 12, 2026version currentsource upstream spec ↗

field	recorded source
schema	Source linked above. upstream-spec Source trace `04c9d2874bad4d95506ccffd28ab3bdbf9fee45bfef4e95e60c65f9656c45c20` { "fetchedAt": 1789189866, "extractorVersion": "1", "verifiedAt": null }

Response body

7 properties · 5 required · 9 $defs

property	type	description	constraints
id*	string	A unique identifier of the response
created*	integer	The Unix time in seconds when the response was generated
model*	string	The model used for the completion
choices*	Choice[]	The list of generated completion choices
usage*	UsageInfo	Usage statistics for the completion
object	string	The object type, which is always "text_completion"	default: "text_completion"
perf_metrics	object	See parameter [perf_metrics_in_response](#body-perf-metrics-in-response)	nullable

Types

9 $defs

Choiceobject

A completion choice.

index*	integer	The index of the completion choice
text*	string	The completion response
logprobs	LogProbs \| NewLogProbs	The log probabilities of the most likely tokens	nullable · default: null
finish_reason	enum	The reason the model stopped generating tokens. This will be "stop" if the model hit a natural stop point or a provided stop sequence, or "length" if the maximum number of tokens specified in the request was reached	stop \| length \| error · nullable
raw_output	RawOutput	Extension of OpenAI that returns low-level interaction of what the model sees, including the formatted prompt and function calls	nullable
prompt_token_ids	integer[]	Token IDs for the prompt (when return_token_ids=true)	nullable
token_ids	integer[]	Token IDs for the generated completion (when return_token_ids=true)	nullable

CompletionTokensDetailsobject

reasoning_tokens

integer

nullable

LogProbsobject

Legacy log probabilities format

tokens	string[]
token_logprobs	number[]
top_logprobs	object[]	nullable
text_offset	integer[]
token_ids	integer[]	nullable

NewLogProbsobject

OpenAI-compatible log probabilities format

content

NewLogProbsContent[]

NewLogProbsContentobject

token*	string
logprob*	number
sampling_logprob*	number	nullable
sampling_mask_count	integer	nullable
sampling_mask	integer[] \| string	nullable · default: null
bytes*	integer[]
top_logprobs	NewLogProbsContentTopLogProbs[]
token_id*	integer
text_offset*	integer
last_activation	string	nullable
routing_matrix	string	nullable

NewLogProbsContentTopLogProbsobject

token*	string
logprob*	number
token_id*	integer
bytes	integer[]

PromptTokensDetailsobject

cached_tokens

integer

nullable

RawOutputobject

Extension of OpenAI that returns low-level interaction of what the model sees, including the formatted prompt and function calls

prompt_fragments*	string \| integer[]	Pieces of the prompt (like individual messages) before truncation and concatenation. Depending on prompt_truncate_len some of the messages might be dropped. Contains a mix of strings to be tokenized and individual tokens (if dictated by the conversation templa…
prompt_token_ids*	integer[]	Fully processed prompt as seen by the model
completion*	string	Raw completion produced by the model before any tool calls are parsed
completion_token_ids	integer[]	Token IDs for the raw completion	nullable
completion_logprobs	NewLogProbs	OpenAI-compatible log probabilities format	nullable
images	string[]	Images in the prompt	nullable
videos	string[]	Videos in the prompt	nullable
grammar	string	Grammar used for constrained decoding, can be either user provided (directly or JSON schema) or inferred by the chat template	nullable

UsageInfoobject

Usage statistics.

prompt_tokens*	integer	The number of tokens in the prompt
total_tokens*	integer	The total number of tokens used in the request (prompt + completion)
completion_tokens	integer	The number of tokens in the generated completion	nullable
prompt_tokens_details	PromptTokensDetails	Details about prompt tokens, including cached tokens	nullable
completion_tokens_details	CompletionTokensDetails	Details about completion tokens, including reasoning tokens	nullable
output_tokens_details	CompletionTokensDetails	Details about output tokens, including reasoning tokens	nullable

Validate a payload

before you spend tokens

shell

$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"fireworks","endpointId":"v1/completions","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}