Schema Reference

The following request / response data structures are auto-synced from newapi’s OpenAPIOpenAPIAn open specification for describing REST APIs (formerly Swagger). Lurus data structures are auto-generated from the gateway's OpenAPI spec.Learn more → spec (relay.json), always staying consistent with the gateway implementation. See the full endpoint list in the API Overview, and call examples in Chat Completions.

How to update

The <!-- sync:schemas --> marker region on this page is regenerated by bun run sync from 2b-svc-newapi/docs/openapi/relay.json. Do not edit this region manually.

35 data structures in total (auto-generated from the OpenAPI spec).

AudioTranscriptionResponse

Field	Type	Required	Description
text	string	—

ChatCompletionRequest

Field	Type	Required	Description
model	string	Yes	Model ID
messages	array<Message>	Yes	List of conversation messages
temperature	number	—	Sampling temperature
top_p	number	—	Nucleus sampling parameter
n	integer	—	Number of generations
stream	boolean	—	Whether to stream the response
stream_options	object	—
stop	string | array<string>	—	Stop sequences
max_tokens	integer	—	Maximum number of tokens to generate
max_completion_tokens	integer	—	Maximum number of completion tokens
presence_penalty	number	—
frequency_penalty	number	—
logit_bias	object	—
user	string	—
tools	array<Tool>	—
tool_choice	string (none / auto / required) | object	—
response_format	ResponseFormat	—
seed	integer	—
reasoning_effort	string (low / medium / high)	—	Reasoning intensity (for models that support reasoning)
modalities	array<string (text / audio)>	—
audio	object	—

ChatCompletionResponse

Field	Type	Required	Description
id	string	—
object	string	—
created	integer	—
model	string	—
choices	array<object>	—
usage	Usage	—
system_fingerprint	string	—

ClaudeMessage

Field	Type	Required	Description
role	string (user / assistant)	Yes
content	string | array<object>	Yes

ClaudeRequest

Field	Type	Required	Description
model	string	Yes
messages	array<ClaudeMessage>	Yes
system	string | array<object>	—
cache_control	object	—
inference_geo	string	—
max_tokens	integer	Yes
temperature	number	—
top_p	number	—
top_k	integer	—
stream	boolean	—
stop_sequences	array<string>	—
tools	array<object>	—
tool_choice	object	—
thinking	object	—
context_management	object	—
output_config	object	—
output_format	object	—
container	string | object	—
mcp_servers	array<object>	—
metadata	object	—
speed	string (standard / fast)	—
service_tier	string (auto / standard_only)	—

ClaudeResponse

Field	Type	Required	Description
id	string	—
type	string	—
role	string	—
content	array<object>	—
model	string	—
stop_reason	string (end_turn / max_tokens / stop_sequence / tool_use)	—
usage	object	—

CompletionRequest

Field	Type	Required	Description
model	string	Yes
prompt	string | array<string>	Yes
max_tokens	integer	—
temperature	number	—
top_p	number	—
n	integer	—
stream	boolean	—
stop	string | array<string>	—
suffix	string	—
echo	boolean	—

CompletionResponse

Field	Type	Required	Description
id	string	—
object	string	—
created	integer	—
model	string	—
choices	array<object>	—
usage	Usage	—

EmbeddingRequest

Field	Type	Required	Description
model	string	Yes
input	string | array<string>	Yes	Text to embed
encoding_format	string (float / base64)	—
dimensions	integer	—	Output vector dimensions

EmbeddingResponse

Field	Type	Required	Description
object	string	—
data	array<object>	—
model	string	—
usage	object	—

GeminiModelsResponse

Field	Type	Required	Description
models	array<object>	—

GeminiResponse

Field	Type	Required	Description
candidates	array<object>	—
usageMetadata	object	—

ImageResponse

Field	Type	Required	Description
created	integer	—
data	array<object>	—

Message

Field	Type	Required	Description
role	string (system / user / assistant / tool / developer)	Yes	Message role
content	string | array<MessageContent>	Yes	Message content
name	string	—	Sender name
tool_calls	array<ToolCall>	—
tool_call_id	string	—	Tool call ID (for tool-role messages)
reasoning_content	string	—	Reasoning content

MessageContent

Field	Type	Required	Description
type	string (text / image_url / input_audio / file / video_url)	—
text	string	—
image_url	object	—
input_audio	object	—
file	object	—
video_url	object	—

Model

Field	Type	Required	Description
id	string	—	Model ID
object	string	—	Object type
created	integer	—	Creation timestamp
owned_by	string	—	Model owner

ModelsResponse

Field	Type	Required	Description
object	string	—
data	array<Model>	—

ModerationRequest

Field	Type	Required	Description
input	string | array<string>	Yes
model	string	—

ModerationResponse

Field	Type	Required	Description
id	string	—
model	string	—
results	array<object>	—

RerankRequest

Field	Type	Required	Description
model	string	Yes
query	string	Yes	Query text
documents	array<string | object>	Yes	List of documents to rerank
top_n	integer	—	Return the top N results
return_documents	boolean	—

RerankResponse

Field	Type	Required	Description
id	string	—
results	array<object>	—
meta	object	—

ResponseFormat

Field	Type	Required	Description
type	string (text / json_object / json_schema)	—
json_schema	object	—	JSON Schema definition

ResponsesCompactionRequest

Field	Type	Required	Description
model	string	Yes
input	string | array<object>	—	Input content, either a string or an array of messages
instructions	string	—
previous_response_id	string	—

ResponsesCompactionResponse

Field	Type	Required	Description
id	string	—
object	string	—
created_at	integer	—
output	array<object>	—
usage	Usage	—
error	object	—

ResponsesRequest

Field	Type	Required	Description
model	string	Yes
input	string | array<object>	—	Input content, either a string or an array of messages
instructions	string	—
max_output_tokens	integer	—
temperature	number	—
top_p	number	—
stream	boolean	—
tools	array<object>	—
tool_choice	string | object	—
reasoning	object	—
previous_response_id	string	—
truncation	string (auto / disabled)	—

ResponsesResponse

Field	Type	Required	Description
id	string	—
object	string	—
created_at	integer	—
status	string (completed / failed / in_progress / incomplete)	—
model	string	—
output	array<object>	—
usage	Usage	—

SpeechRequest

Field	Type	Required	Description
model	string	Yes
input	string	Yes	Text to convert
voice	string (alloy / echo / fable / onyx / nova / shimmer)	Yes
response_format	string (mp3 / opus / aac / flac / wav / pcm)	—
speed	number	—

Tool

Field	Type	Required	Description
type	string	—
function	object	—

ToolCall

Field	Type	Required	Description
id	string	—
type	string	—
function	object	—

Usage

Field	Type	Required	Description
prompt_tokens	integer	—	Number of prompt tokens
completion_tokens	integer	—	Number of completion tokens
total_tokens	integer	—	Total number of tokens
prompt_tokens_details	object	—
completion_tokens_details	object	—

VideoRequest

Video generation request

Field	Type	Required	Description
model	string	—	Model/style ID
prompt	string	—	Text description prompt
image	string	—	Image input (URL or Base64)
duration	number	—	Video duration (seconds)
width	integer	—	Video width
height	integer	—	Video height
fps	integer	—	Video frame rate
seed	integer	—	Random seed
n	integer	—	Number of videos to generate
response_format	string	—	Response format
user	string	—	User identifier
metadata	object	—	Extended parameters (e.g. negative_prompt, style, quality_level)

VideoResponse

Video generation task submission response

Field	Type	Required	Description
task_id	string	—	Task ID
status	string	—	Task status

VideoTaskError

Video task error information

Field	Type	Required	Description
code	integer	—	Error code
message	string	—	Error message

VideoTaskMetadata

Video task metadata

Field	Type	Required	Description
duration	number	—	Actual generated video duration
fps	integer	—	Actual frame rate
width	integer	—	Actual width
height	integer	—	Actual height
seed	integer	—	Random seed used

VideoTaskResponse

Video task status query response

Field	Type	Required	Description
task_id	string	—	Task ID
status	string (queued / in_progress / completed / failed)	—	Task status
url	string	—	Video resource URL (on success)
format	string	—	Video format
metadata	VideoTaskMetadata	—
error	VideoTaskError	—

Next Steps

API OverviewChat CompletionsIntegrations & MCP DirectoryError Handling