FLUX 3 Dev API: Python Wrapper for Black Forest Labs' Fast, Low-Cost FLUX 3 Variant

A Python wrapper for FLUX 3 Dev — the faster, lower-cost variant of Black Forest Labs' FLUX 3 frontier model — delivered via muapi.ai. FLUX 3 Dev trades a small amount of peak fidelity for significantly reduced latency and cost, making it ideal for rapid iteration, prototyping, and high-volume generation, with an open-weight release planned for later in 2026. This SDK also covers the full FLUX 3 family through the same client.

⚠️ Status: FLUX 3 Text-to-Video and Image-to-Video are live now on MuAPI. FLUX 3 Dev, flagship Text-to-Image, and Image-to-Image (the three endpoints this repo is named after) are still coming soon — they're wired into this SDK ahead of launch, but the underlying MuAPI routes aren't enabled yet and calls to generate()/text_to_image()/image_to_image() will fail until Black Forest Labs opens general availability. See muapi.ai/flux-3 for current status of every FLUX 3 endpoint.

MuAPI's FLUX 3 video API has also expanded beyond Text-to-Video/Image-to-Video since this SDK was written — Start-End-to-Video, Video Extend, Video Upscaler, and fast draft modes are live too. This repo doesn't wrap those; see flux-3-video-api (all seven live video endpoints), flux-3-video-edit (Video Extend), and flux-3-video-upscaler if you need those.

📺 FLUX 3 API: Image & Video Generation from One Multimodal Model →

Related Projects

• MiniMax-H3-API — Python SDK for MiniMax H3 text-to-video, image-to-video, and first/last-frame video generation.
• Wan-3.0-API — Python SDK and MCP server for Wan 3.0-compatible AI video generation.
• awesome-flux-3-api-prompts — FLUX 3 API guide, prompt engineering, and a curated prompt library
• flux-3-video-api — Python wrapper covering all seven live FLUX 3 video endpoints (Text-to-Video, Image-to-Video, Start-End-to-Video, Video Extend, Video Upscaler, plus draft modes)
• flux-3-video-edit — Python SDK for FLUX 3 Video Extend (continues an existing clip with new prompt-guided motion)
• flux-3-video-upscaler — Python SDK for upscaling FLUX 3 (or any) video output beyond its native resolution
• flux-3-omni — Python SDK for FLUX 3's multi-reference Omni Reference mode (coming soon)
• Seedance-2-API — Python SDK for ByteDance's Seedance 2.x video models
• awesome-ai-image-models — comparison of AI image generation models across providers
• Grok-Imagine-Image-2-API — Python SDK and MCP server for Grok Imagine Image 2.0 image generation and editing through MuAPI
• Open-Generative-AI — open-source, self-hosted AI image & video generation studio (200+ models)
• ai-creator-academy — free curriculum teaching creators how to monetize FLUX and other generative AI models

🚀 Why Use FLUX 3 Dev?

FLUX 3 is Black Forest Labs' newest frontier model — a unified multimodal system jointly trained across image, video, and audio, extendable to action prediction for robotics. FLUX 3 Dev is its fast, affordable entry point:

• Lower Latency & Cost: Trades a small amount of peak fidelity from the flagship model for significantly faster, cheaper generation.
• Same Prompting Conventions: Uses the same prompt structure as flagship FLUX 3 — validate ideas cheaply, then re-run on the full model if needed.
• Planned Open-Weight Release: Black Forest Labs has confirmed a faster, open-weight version of FLUX 3 is coming later in 2026.
• Full FLUX 3 Family Access: This SDK also wraps FLUX 3 Text-to-Image, Image-to-Image, Text-to-Video, and Image-to-Video — one client for the whole model family (see status note above — Dev/T2I/I2I are coming soon, T2V/I2V are live).
• Native Audio in Video: FLUX 3 Video variants can generate scene-appropriate synchronized audio directly alongside the clip, no extra charge.
• Developer-First: Simple Python SDK on top of MuAPI's unified infrastructure — no separate account or waitlist needed once you have a MuAPI key.

🌟 Key Features

• 🕒 FLUX 3 Dev Text-to-Image (coming soon): Fast, low-cost image generation via generate().
• 🕒 FLUX 3 Text-to-Image (coming soon): Flagship higher-fidelity image generation via text_to_image().
• 🕒 FLUX 3 Image-to-Image (coming soon): Instruction-driven editing with up to 4 reference images via image_to_image().
• ✅ FLUX 3 Text-to-Video (live): Cinematic video with optional native synchronized audio via text_to_video().
• ✅ FLUX 3 Image-to-Video (live): Animate a still image into video with optional native audio via image_to_video().
• ✅ File Upload: Upload local images/videos directly using upload_file().
• ✅ MCP Server: Use FLUX 3 Dev as a Model Context Protocol server for Claude Desktop, Cursor, and other MCP clients.

---

🛠 Installation

Via Pip (Recommended)

pip install flux-3-dev-api

From Source

git clone https://github.com/Anil-matcha/Flux-3-Dev-API.git
cd Flux-3-Dev-API
pip install -r requirements.txt

Configuration

Create a .env file in the root directory and add your MuAPI API key:

MUAPI_API_KEY=your_muapi_api_key_here

---

🤖 FLUX 3 Dev MCP Server

Use FLUX 3 Dev as an MCP (Model Context Protocol) server so AI clients (like Claude Desktop or Cursor) can directly invoke FLUX 3 generation tools.

Running the MCP Server

1. Ensure MUAPI_API_KEY is set in your environment.
2. Run the server:

   python3 mcp_server.py
   

3. To test with the MCP Inspector:

   npx -y @modelcontextprotocol/inspector python3 mcp_server.py
   

---

💻 Quick Start (Python)

from flux3_dev_api import Flux3DevAPI

# Initialize the client
api = Flux3DevAPI()

# Generate an image with FLUX 3 Dev
print("Generating image with FLUX 3 Dev...")
submission = api.generate(
    prompt="A minimalist product shot of a ceramic mug on a wooden table, soft natural light",
    aspect_ratio="1:1",
    resolution="1k"
)

# Wait for completion
result = api.wait_for_completion(submission["request_id"])
print(f"Success! Output: {result['outputs'][0]}")

---

📡 API Endpoints & Reference

1. FLUX 3 Dev (Text-to-Image, fast/low-cost) — 🕒 Coming Soon

Endpoint: POST https://api.muapi.ai/api/v1/flux-3-dev

curl --location --request POST "https://api.muapi.ai/api/v1/flux-3-dev" \
  --header "Content-Type: application/json" \
  --header "x-api-key: YOUR_API_KEY" \
  --data-raw '{
      "prompt": "A minimalist product shot of a ceramic mug on a wooden table, soft natural light",
      "aspect_ratio": "1:1",
      "resolution": "1k"
  }'

Python SDK:

submission = api.generate(
    prompt="A minimalist product shot of a ceramic mug on a wooden table, soft natural light",
    aspect_ratio="1:1",
    resolution="1k",
)
result = api.wait_for_completion(submission["request_id"])
print(result["outputs"][0])

2. FLUX 3 Text-to-Image (flagship) — 🕒 Coming Soon

Endpoint: POST https://api.muapi.ai/api/v1/flux-3-text-to-image

curl --location --request POST "https://api.muapi.ai/api/v1/flux-3-text-to-image" \
  --header "Content-Type: application/json" \
  --header "x-api-key: YOUR_API_KEY" \
  --data-raw '{
      "prompt": "A hyperrealistic portrait of an astronaut standing on red Martian dunes at golden hour",
      "aspect_ratio": "16:9",
      "resolution": "2k"
  }'

Python SDK:

submission = api.text_to_image(
    prompt="A hyperrealistic portrait of an astronaut standing on red Martian dunes at golden hour",
    aspect_ratio="16:9",
    resolution="2k",
)

3. FLUX 3 Image-to-Image (Edit) — 🕒 Coming Soon

Endpoint: POST https://api.muapi.ai/api/v1/flux-3-image-to-image

curl --location --request POST "https://api.muapi.ai/api/v1/flux-3-image-to-image" \
  --header "Content-Type: application/json" \
  --header "x-api-key: YOUR_API_KEY" \
  --data-raw '{
      "prompt": "Replace the background with a neon-lit city street at night, keep the subject unchanged",
      "images_list": ["https://example.com/input.jpg"],
      "aspect_ratio": "1:1"
  }'

Python SDK:

submission = api.image_to_image(
    prompt="Replace the background with a neon-lit city street at night, keep the subject unchanged",
    images_list=["https://example.com/input.jpg"],
    aspect_ratio="1:1",
)

4. FLUX 3 Text-to-Video — ✅ Live

Endpoint: POST https://api.muapi.ai/api/v1/flux-3-text-to-video · $0.25/sec @720p, $0.42/sec @1080p

curl --location --request POST "https://api.muapi.ai/api/v1/flux-3-text-to-video" \
  --header "Content-Type: application/json" \
  --header "x-api-key: YOUR_API_KEY" \
  --data-raw '{
      "prompt": "A drone shot glides over a bioluminescent forest at night, fireflies drifting between glowing trees",
      "aspect_ratio": "16:9",
      "resolution": "1080p",
      "duration": 6,
      "generate_audio": true
  }'

Python SDK:

submission = api.text_to_video(
    prompt="A drone shot glides over a bioluminescent forest at night, fireflies drifting between glowing trees",
    aspect_ratio="16:9",
    resolution="1080p",
    duration=6,
    generate_audio=True,
)

5. FLUX 3 Image-to-Video — ✅ Live

Endpoint: POST https://api.muapi.ai/api/v1/flux-3-image-to-video · $0.25/sec @720p, $0.42/sec @1080p

curl --location --request POST "https://api.muapi.ai/api/v1/flux-3-image-to-video" \
  --header "Content-Type: application/json" \
  --header "x-api-key: YOUR_API_KEY" \
  --data-raw '{
      "prompt": "The camera slowly pushes in as steam rises from the coffee cup",
      "image_url": "https://example.com/product.jpg",
      "duration": 5,
      "generate_audio": true
  }'

Python SDK:

submission = api.image_to_video(
    prompt="The camera slowly pushes in as steam rises from the coffee cup",
    image_url="https://example.com/product.jpg",
    duration=5,
    generate_audio=True,
)

---

🎬 Other Live FLUX 3 Video Endpoints

This SDK only wraps Text-to-Video and Image-to-Video. MuAPI's FLUX 3 video API also includes these live endpoints, covered by sibling projects:

---

📖 Method Reference

---

🔗 Official Resources

• FLUX 3 Announcement (Black Forest Labs): bfl.ai/models/flux-3
• Playground — FLUX 3 (all variants): muapi.ai/flux-3
• Playground — FLUX Kontext (available today): muapi.ai/flux-kontext
• API Provider: MuAPI.ai

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

---

Keywords: FLUX 3 API, FLUX 3 Dev, FLUX 3 Dev API, Black Forest Labs FLUX 3, FLUX 3 Python SDK, FLUX 3 text to image, FLUX 3 image to image, FLUX 3 text to video, FLUX 3 image to video, FLUX multimodal model, AI image generation API, AI video generation API, MuAPI, Python image generation SDK, open-weight FLUX.