Provider Hub

Turn set descriptions and reference images into 3D environments for Scene Builder. Connect your personal World Labs account; generation uses your own API credits.

• spatial

Needs: API Key

Optional direct OpenAI routing for GPT-6. Connect this only if you prefer OpenAI rather than using GPT-6 through an already-connected Muapi.ai account.

• Text
• Reasoning
• Image
• Multimodal
• Code
• Embeddings

Needs: API Key · Organization ID (optional)

Google's Gemini family offers exceptional multimodal understanding — analyzing script PDFs, images, and video together. The ultra-long context window makes it ideal for full screenplay analysis and world-building.

• Text
• Multimodal
• Long context
• Reasoning
• Image

Needs: Gemini API Key

Run Google's open Gemma 4 models through the hosted Gemini API — no local GPU required, with long context. A budget-friendly engine for Aria, screenplay analysis, brainstorming, and structured writing. Uses your Gemini API key. Gemma powers Aria and text-based AI functions only — image and video generation continue to use your configured media providers.

• Text
• Multimodal
• Long context
• Reasoning

Needs: Gemini API Key

Connect any OpenAI-compatible API — LM Studio, private deployments on vLLM, or any hosted endpoint that follows the OpenAI chat completions spec.

• Text

Needs: Base URL · API Key (optional) · Default Model ID

Connect your Qwen API for writing and reasoning.

• Text
• Reasoning
• Long context

Needs: API Key · Base URL (optional) · Model name

Connect DeepSeek directly — separate from the MuApi DeepSeek model.

• Text
• Reasoning

Needs: API Key · Base URL (optional) · Model name

Gemma 4 from Google DeepMind — state-of-the-art open-weight language models. Available in Dense (31B) and MoE (26B) architectures. Run locally via Ollama or connect via any OpenAI-compatible endpoint. Gemma powers Aria and text-based AI functions only — image and video generation continue to use your configured media providers.

• Text
• Reasoning
• Multimodal
• Fast
• Budget

Needs: Base URL · API Key (optional)

Muapi.ai is a unified API gateway for image, video, and (when your account exposes it) GPT-6 intelligence. One key can cover media generation and advanced language models — you do not need a separate OpenAI account for GPT-6 if Muapi already provides it.

• Text
• Image
• Video
• Image→Video
• Text→Video

Needs: API Key

Google's Nano Banana 2 (Gemini 3.1 Flash Image) produces high-fidelity images up to 4K with exceptional text rendering and subject consistency.

• Image
• Multimodal
• Fast

Needs: Gemini API Key

Run ComfyUI workflows — including comfy.org Partner/API nodes — from your own ComfyUI server. Your comfy.org account API key authenticates those API nodes; it is injected into the workflow request's extra_data (api_key_comfy_org), not an Authorization header. Point it at a local install (http://127.0.0.1:8188) or a hosted ComfyUI server.

• Image
• Video
• Image→Video

Needs: comfy.org API Key · ComfyUI Server URL

ElevenLabs produces the most natural-sounding AI voice synthesis available. Essential for creating temp tracks, ADR scratch audio, character voice demos, and audio production assets directly within your creative pipeline.

• Audio
• Premium

Needs: API Key