Flux 2 Turbo Text To Image API Documentation

Flux 2 Turbo Text To Image API Documentation

Playground

FLUX 2 turbo from Black Forest Labs is the speed-optimized text-to-image model for real-time workflows. Generate photoreal images and clean typography with strong prompt adherence and consistent style—ideal for ads, posters, social posts, and rapid iteration. Built for low-latency, high-throughput use. Ready-to-use REST API, best performance, no cold starts, affordable pricing.

Features

FLUX.2 [turbo] is the speed-optimized text-to-image model in the FLUX.2 family, built for ultra-fast generation with strong prompt adherence and dependable output quality. It’s ideal for real-time creation, high-volume pipelines, and rapid iteration where latency matters.

Where FLUX.2 [turbo] fits best

• Real-time applications requiring low latency
• High-volume batch generation workflows
• Rapid iteration, prototyping, and creative exploration
• Cost-sensitive production pipelines that still need consistent quality

Key benefits

• Turbo-fast generation Optimized for minimal latency so you can generate more variations per minute.

• Quality you can ship Designed to preserve the core FLUX.2 look and coherence while prioritizing speed.

• Prompt-smart outputs Handles detailed prompts (objects, lighting, style cues) with reliable composition and fewer “random surprises.”

• Format-ready files Supports common output formats including JPEG, PNG, and WebP for web, design, and production workflows.

Parameters

How to use

1. Write a detailed prompt describing subject, environment, style, and lighting.
2. Set output size by choosing width and height (square for general use, wide for banners, tall for posters).
3. Set a seed:

• Use -1 for random variations each run
• Use a fixed number to reproduce a result or explore controlled variations

4. Run the model and download the generated image.

Pricing

• $0.04 per generated image

FLUX.2 family on WaveSpeedAI

• FLUX.2 [turbo] Edit – ultra-fast image-to-image editing.
• FLUX.2 [flash] Text-to-Image – lightning-fast text-to-image generation.
• FLUX.2 [pro] Text-to-Image – premium quality text-to-image.
• Flux-2-dev/text-to-image — Lightweight FLUX.2 text-to-image generation optimized for speed, large-batch runs, and LoRA training workflows.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "prompt": "A cinematic ocean wave at sunrise, highly detailed",
  "size": "1024*1024"
}
JSON
)

# 1. Submit the prediction once. Never retry this request automatically:
# a retry after an unclear failure can create a second, billed task.
CHECK='The task may already exist. Check History or the predictions list before submitting again.'
if ! SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  --connect-timeout 10 --max-time 60 \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/flux-2-turbo/text-to-image" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}"); then
  printf 'Submit failed. %s\n' "${CHECK}" >&2
  exit 1
fi

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id. %s\n' "${CHECK}" >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 2. Poll until the prediction reaches a final status.
BODY_FILE=$(mktemp)
trap 'rm -f "${BODY_FILE}"' EXIT
while true; do
  # Transient errors (429, 5xx) are retried, honoring Retry-After.
  HTTP_CODE=$(curl --silent --show-error --fail-with-body \
    --connect-timeout 10 --max-time 60 --retry 5 --retry-delay 2 \
    --output "${BODY_FILE}" --write-out '%{http_code}' \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}" || true)
  case "${HTTP_CODE}" in
    2??) ;;
    000|429|5??)  # dropped connection or temporary error: keep polling
      printf 'Task %s: HTTP %s, polling again in 10s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      sleep 10; continue ;;
    *) printf 'Task %s: HTTP %s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      cat "${BODY_FILE}" >&2; exit 1 ;;
  esac
  RESULT=$(jq 'if type == "object" and has("data") then .data else . end' "${BODY_FILE}")
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted)
      printf 'Task %s ended with status %s\n' "${PREDICTION_ID}" "${STATUS}" >&2
      printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Every model also accepts the global request options enable_sync_mode and enable_base64_output; see their pages for details.

Response Parameters

Result Request Parameters

Result Response Parameters