Skip to main content
POST
chat completions

Maven

Normal Request Example

Normal Response

Stream Response

Body

application/json
messages
any[]
required

messages

Minimum array length: 1
model
enum<string>
required

model, You can view the list of supported models through the interface /ai/chat/models

Available options:
gpt-5.4,
gpt-5.5,
claude-opus-4-6,
claude-opus-4-7,
claude-opus-4-8
stream
boolean
default:false

streaming output

temperature
number<double>

Temperature (0-2)

maxTokens
integer<int32>

Maximum token count

maxCompletionTokens
integer<int32>

Maximum number of completed tokens

topP
number<double>

top P

frequencyPenalty
number<double>

frequency penalty(-2.0 to 2.0)

presencePenalty
number<double>

presence penalty(-2.0 to 2.0)

stop
any

Stop sequence

n
integer<int32>

Generated Quantity

seed
integer<int32>

random seed

user
string

user

tools
any[]

Tool List (Function Calling)

toolChoice
any

Tool selection strategy

responseFormat
any

Response format

logprobs
boolean

logprobs

topLogprobs
integer<int32>

top logprobs

logitBias
any

logit bias

parallelToolCalls
boolean

parallel tool calls

Response

200 - */*

OK

The response is of type object.