fal-ai-media

Generate images, videos, and audio using fal.ai models via MCP.

Updated Mar 26, 2026
One-click install
npx skills add https://github.com/inuishan/PET --skill fal-ai-media-inuishan
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: fal-ai-media
Source: https://github.com/inuishan/PET/tree/main/.codex/skills/fal-ai-media
Command: npx skills add https://github.com/inuishan/PET --skill fal-ai-media-inuishan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires fal-ai-mcp-server.

What problem does it solve? Creating AI-generated media requires knowing which model to use, its parameters, and how to call it. This Skill provides ready-to-use invocation patterns for fal.ai's image, video, and audio models through a single MCP interface, removing the guesswork from model selection and parameter configuration. ## Core Features & Use Cases - Image Generation: Text-to-image and image editing with Nano Banana 2 for fast drafts and Nano Banana Pro for high-fidelity production output. - Video Generation: Text-to-video and image-to-video using Seedance 1.0 Pro, Kling Video v3 Pro, and Veo 3, with duration, aspect ratio, and seed controls. - Audio Generation: Conversational text-to-speech with CSM-1B, video-to-audio with ThinkSound, plus ElevenLabs and VideoDB alternatives for voice, music, and sound effects. - Use Case: A content creator needs a thumbnail, a short promo clip, and a voiceover. Use this Skill to generate the image with Nano Banana Pro, animate it with Seedance, and synthesize narration with CSM-1B, checking cost estimates before each run. ## Quick Start Ask the AI to generate a 16:9 landscape image of your chosen subject using the fal.ai Nano Banana 2 model.

Frequently Asked Questions about fal-ai-media

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate images with fal.ai?▼

Call the generate MCP tool with model_name fal-ai/nano-banana-2 for fast drafts or fal-ai/nano-banana-pro for high-fidelity output. Provide a prompt, image_size such as landscape_16_9, and optionally num_images, seed, and guidance_scale.

How to create a video from an image using AI?▼

Upload the source image with the upload tool, then call generate with fal-ai/seedance-1-0-pro, passing the uploaded image_url, a motion-focused prompt, duration of 5s or 10s, and an aspect ratio. Image-to-video gives more controlled results than pure text-to-video.

What fal.ai model should I use for text-to-video?▼

Seedance 1.0 Pro suits high motion quality, Kling Video v3 Pro adds native audio generation, and Veo 3 produces video with generated sound and high visual quality. Use the search or find MCP tools to compare models for your task.

Does fal.ai media generation require an API key?▼

Yes, the fal.ai MCP server requires a FAL_KEY environment variable configured in ~/.claude.json. You can obtain an API key from fal.ai and add it to the server configuration before invoking any generation tools.

How can I estimate fal.ai generation cost before running?▼

Use the estimate_cost MCP tool with the model name and input parameters before generating. This is recommended especially for expensive video generations, and you can iterate on prompts with lower-cost models like Nano Banana 2 first.