Stream object

Stream a structured response and display fields as soon as they appear with streamObject.

The final response is decoded into Recipe, while onPartial receives the growing JSON tree. Partial values are useful for progress displays, but use the final recipe.value for validated application data.

import std/[json, os, strutils]
import nimgent
import nimgent/providers/openai

type Recipe = object
  name: string
  servings: int
  ingredients: seq[string]

let model = openAI(getEnv("OPENAI_API_KEY")).model("gpt-4o-mini")

echo "calling the model..."
let recipe = streamObject[Recipe](
  model,
  prompt = "A weeknight lasagna.",
  onPartial = proc (partial: JsonNode): bool =
    if "name" in partial:
      stdout.write "\rname: " & partial["name"].getStr
      flushFile(stdout)
    true)

echo ""
echo "name: ", recipe.value.name
echo "servings: ", recipe.value.servings
echo "ingredients: ", recipe.value.ingredients.join(", ")

Run it from the repository root:

OPENAI_API_KEY=... nim c -r examples/stream_object.nim

View the source example