Run Specific Prompt Version
Renders a specific template version of a prompt and executes it against the configured model provider.
Authorizations
Use an access token retrieved from login (Bearer sess_...) or a programmatic API key (Bearer ak_...).
Path Parameters
The ID of the project the prompt belongs to.
Prompt slug, unique within the project.
Version sequence number (integer) or version tag (string) to execute.
Body
Variables to interpolate into the template.
Set to true to receive token stream instead of a blocked response.
Ad-hoc model override. Must start with a supported provider prefix: 'openai/', 'anthropic/', 'gemini/', 'deepseek/', 'groq/', or 'openrouter/'. All providers except 'anthropic/' follow standard OpenAI-compatible JSON request, response, and SSE streaming conventions.
Ad-hoc model parameters override.
Response
Completed response text or stream chunks (Streaming SSE events where each line is formatted as data: {"delta": string, "done": boolean}).
Text content produced by the model provider completion.
"Greetings from the model!"
The model identifier that was executed.
"openai/gpt-4o"
The prompt version number that was used.
2
Prompt slug.
"my-prompt"

