Skip to main content
POST
/
v1
/
llm
/
generations
curl --request POST \
  --url https://api.foxapi.cc/v1/llm/generations \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "gemini-2.5-pro",
  "audio_url": "https://storage.googleapis.com/cloud-samples-tests/speech/brooklyn.flac",
  "max_tokens": 256
}
'
{
  "id": "task-llm-1776874481-rj6bs3yb",
  "object": "llm.generation.task",
  "type": "llm",
  "model": "gemini-2.5-pro",
  "status": "pending",
  "progress": 0,
  "created": 1776874481,
  "stream": null,
  "results": null,
  "error": null
}

Authorizations

​
Authorization
string
header
required

All endpoints require Bearer Token authentication. Add to the request header:

Authorization: Bearer YOUR_API_KEY

YOUR_API_KEY is the API Token (sk-... format).

Body

application/json
​
model
string
required

Get Model List

Example:

"gemini-2.5-pro"

​
audio_url
string
required

Audio source. Accepts one of the following two forms:

  • Publicly reachable HTTP/HTTPS URL
  • data:audio/<type>;base64,<payload> data URI (base64 inline)

Audio format support per family (the specific available models are driven by channel configuration):

  • Gemini family (e.g. gemini-*): wav/mp3/aiff/aac/ogg/flac/m4a; total request body (prompt + system + inline files) ≤ 20 MB

Base64 data is not size-validated; oversized payloads may trigger 422.

Minimum string length: 1
Example:

"https://storage.googleapis.com/cloud-samples-tests/speech/brooklyn.flac"

​
prompt
string | null

User prompt. When omitted, defaults to 'Please transcribe this audio file', aligning with the transcription scenario.

Maximum string length: 100000
Example:

"Identify the speakers and emotion in this audio."

​
sync
boolean
default:false

Synchronous mode. When true, the endpoint blocks until the upstream completes and returns the full response (if stream=true at the same time, returns an SSE stream); when false, the endpoint returns the task ID immediately, and results are fetched via GET /v1/tasks/{task_id} or the SSE endpoint.

Example:

false

​
stream
boolean
default:false

Whether to stream. When true, the Submit response includes stream.url pointing to the SSE subscription path; streaming chunks are unified as the OpenAI chat.completion.chunk format.

Example:

false

​
max_tokens
integer | null

Generation token limit. Optional.

Required range: x >= 1
Example:

256

​
temperature
number | null

Sampling temperature, range [0, 2]. Optional.

Required range: 0 <= x <= 2
​
system_prompt
string | null

System instruction. Optional.

Maximum string length: 10000
​
reasoning
boolean | null

Whether to include reasoning tokens. Some thinking models require this to be set to true.

Response

Task created

Submit response, conforming to the unified task standard shape. results / error are fixed at null during submit; they are returned via GET /v1/tasks/{task_id} after the task completes or fails.

​
id
string
required

Task ID, formatted as task-llm-{timestamp}-{8random}.

Example:

"task-llm-1776874565-yq3szvcu"

​
object
enum<string>
required
Available options:
llm.generation.task
Example:

"llm.generation.task"

​
type
enum<string>
required
Available options:
llm
Example:

"llm"

​
model
string
required

The model name submitted by the client (echoed verbatim)

Example:

"gemini-2.5-pro"

​
status
enum<string>
required
Available options:
pending
Example:

"pending"

​
progress
integer
required
Example:

0

​
created
integer
required
Example:

1776874565

​
stream
object

Returns {url: ...} when stream=true; null when stream=false.

​
results
object[] | null

Fixed at null during submit; returned via GET /v1/tasks/{task_id} after the task completes — results[0] is the full OpenAI ChatCompletion response (audio transcription / understanding output is in message.content).

Example:

null

​
error
object

Fixed at null during submit; returned via GET /v1/tasks/{task_id} when the task fails.

Example:

null