Skip to main content
POST
Create chat response
Use this endpoint to send plain text input to a model and receive a plain text response. This Capriole-native route is non-streaming; use Chat Completions, Responses, or Messages when your client needs an SSE stream. POST /v1/chat accepts openai-latest, claude-latest, google-latest, and public concrete model IDs returned by GET /v1/models. Use a latest alias to let Capriole AI choose our recommended flagship model for that provider. Use a concrete model ID when version pinning matters. Capriole AI web chat and the public API are separate product surfaces. In web chat, Fable 5 and Fable 5 Thinking are the primary Anthropic modes, while Opus 5 and Opus 5 Thinking appear under Other Models. The public API does not expose web chat thinking presets as separate model IDs; for Claude Chat API requests, use claude-latest or a concrete public Claude model ID such as anthropic/claude-fable-5, anthropic/claude-opus-5, or anthropic/claude-sonnet-4-6. Existing Opus 4.8, Opus 4.7, and Opus 4.6 integrations remain supported.

Authorizations

Authorization
string
header
required

Use an API key created in the Capriole AI page. Send it as Authorization: Bearer sk-....

Body

application/json
model
enum<string>
required

Public model identifier or latest alias returned by GET /v1/models

Available options:
openai-latest,
openai/gpt-5.6-terra,
openai/gpt-5.6-luna,
openai/gpt-5.5,
openai/gpt-5.4-mini,
claude-latest,
anthropic/claude-fable-5,
anthropic/claude-opus-5,
anthropic/claude-opus-4-8,
anthropic/claude-opus-4-7,
anthropic/claude-opus-4-6,
anthropic/claude-sonnet-4-6,
google-latest,
google/gemini-3.1-pro-preview,
google/gemini-3.5-flash,
xai/grok-4.5,
zai/glm-5.2,
moonshot/kimi-k3
input
string
required

Plain text user input

Enable provider-native web search when the selected model supports it.

temperature
number

Optional sampling temperature.

Required range: x >= 0
max_output_tokens
integer

Optional maximum number of output tokens.

max_retries
integer

Optional maximum number of provider retries.

Required range: x >= 0
timeout
number

Optional provider request timeout in seconds.

Response

Chat completion response

id
string
required
model
string
required
result
object
required
usage
object
required
Last modified on July 28, 2026