Skip to main content
POST
/
v3
/
roundtable
Roundtable
curl --request POST \
  --url https://api.opper.ai/v3/roundtable \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "input": "<string>",
  "models": [
    {
      "name": "<string>",
      "max_tokens": 123,
      "reasoning": "<string>",
      "temperature": 123,
      "top_p": 123
    }
  ]
}
'
{
  "data": "<string>",
  "id": "<string>",
  "meta": {
    "models_used": [
      "<string>"
    ],
    "resolution": "<string>",
    "total_cost": 123,
    "total_duration_ms": 123,
    "summary_model": "<string>",
    "summary_usage": {
      "input_tokens": 123,
      "output_tokens": 123,
      "cache_creation_1h_tokens": 123,
      "cache_creation_tokens": 123,
      "cache_read_tokens": 123,
      "extra": {},
      "input_audio_tokens": 123,
      "output_audio_tokens": 123,
      "reasoning_tokens": 123,
      "web_fetch_requests": 123,
      "web_search_requests": 123
    },
    "trace_uuid": "<string>"
  },
  "model_results": [
    {
      "data": "<string>",
      "duration_ms": 123,
      "index": 123,
      "model": "<string>",
      "cost": 123,
      "error": "<string>",
      "schema_error": "<string>",
      "schema_valid": true,
      "usage": {
        "input_tokens": 123,
        "output_tokens": 123,
        "cache_creation_1h_tokens": 123,
        "cache_creation_tokens": 123,
        "cache_read_tokens": 123,
        "extra": {},
        "input_audio_tokens": 123,
        "output_audio_tokens": 123,
        "reasoning_tokens": 123,
        "web_fetch_requests": 123,
        "web_search_requests": 123
      }
    }
  ]
}

Documentation Index

Fetch the complete documentation index at: https://docs.opper.ai/llms.txt

Use this file to discover all available pages before exploring further.

This endpoint is in beta. Its request and response shapes may change without the usual deprecation window.

Authorizations

Authorization
string
header
required

API key authentication. Pass your API key as a Bearer token.

Body

application/json
input
string
required

The question or prompt sent to all models

models
object[]
required

Models to query in parallel

choices
string[]

Literal options for multiple_choice resolution

instructions
string

System prompt shared across all model calls

max_retries
integer

Per-model retry count on transient errors (default 1 / max 3)

output_schema
object

JSON Schema for structured output — applied to all model calls and the consolidation call

resolution
string

Resolution mode: summary (default) / fast / multiple_choice

stream
boolean

Enable SSE streaming

summary_model
string

Override the model used for consolidation in summary mode

timeout_ms
integer

Per-model timeout in milliseconds covering all retries (default 120000 / max 300000)

Response

Successful response

data
string
required

Final consolidated output. Null in fast mode (no consolidation runs).

id
string
required

Stable identifier for this roundtable call.

meta
object
required

Aggregated call metadata.

model_results
object[]
required

Per-model fan-out results in the order they appeared in the request.