메뉴
AI model selection guidance for media generation — which image, video, TTS, and music model to use for each use case. Triggers on "which model", "best model for", "image generation model", "video model", "TTS provider", or cost optimization questions.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Content repurposing strategy and extraction rules — how to derive maximum formats from a single generated asset. Triggers on "repurpose", "reuse content", "turn this into", "make a carousel from this video", or when /repurpose command is invoked.
Create a full marketing plan for any project — market research, positioning, funnel strategy, content roadmap, channel mix, and KPIs. Triggers on "marketing plan", "marketing strategy", "go-to-market", "GTM plan", "marketing roadmap", "how to market this", or when a user needs a comprehensive marketing blueprint for a product or service.
Proven ad script frameworks and copywriting patterns for video ads. Triggers on "write ad script", "ad copy", "hook ideas", "curiosity loop", "script structure", or when any /generate-ad or /generate-ugc-ad command is invoked.
Content strategy guidance for social media marketing — category psychology, posting cadence, retention optimization, and content planning. Triggers on "content strategy", "posting schedule", "what to post", "content calendar", "category balance", or engagement optimization questions.
Video production best practices and learnings for AI-generated content. Triggers on "video quality", "production tips", "Seedance tips", "Kling tips", "lip-sync", "ffmpeg composition", or video generation troubleshooting.
| name | gen-ai-media |
| description | AI model selection guidance for media generation — which image, video, TTS, and music model to use for each use case. Triggers on "which model", "best model for", "image generation model", "video model", "TTS provider", or cost optimization questions. |
| Model | Via | Cost | Best For | Quality |
|---|---|---|---|---|
| Ideogram V3 Turbo | Replicate | $0.02/img | General-purpose, stylized images | High, stylized |
| GPT Image 1.5 Low | Replicate | $0.013/img | Product cutouts, simple items | Good, photorealistic |
| GPT Image 1.5 Medium | Replicate | $0.05/img | Compositions, character designs | Very good |
| GPT Image 1.5 High | Replicate | $0.14/img | Hero images, avatar frames | Best, full fidelity |
| FLUX.1 Kontext Pro | Replicate | $0.04/img | Budget avatar frames with reference | Good, fast |
| FLUX.1 Fill Pro | Replicate | $0.05/img | Outpainting / background extension | Seamless extension |
| Nano Banana (Gemini 2.5 Flash) | Google AI | $0.15/img | Alternative composition | Good |
| Bria BG Removal | Replicate | $0.001/img | Background removal | Clean edges |
| Model | Via | Cost | Best For |
|---|---|---|---|
| Seedance 1.5 Pro 480p | Replicate | $0.013/sec (no audio), $0.025/sec (with audio) | Cost-efficient reels |
| Seedance 1.5 Pro 720p | Replicate | $0.026/sec (no audio), $0.052/sec (with audio) | Higher quality |
| Seedance 1 Lite 480p | Replicate | $0.018/sec | Legacy, avoid |
| prunaai/p-video | Replicate | ~$0.07/shot | Audio-driven animation (talking characters) |
| PixVerse Lipsync | Replicate | $0.08325/sec | Post-production lip-sync |
| Provider | Model | Cost | Best For |
|---|---|---|---|
| MiniMax | Speech 2.8 HD | ~$0.000005/char | Budget voiceovers, marketing content |
| ElevenLabs | Turbo v2.5 / Multilingual v2 | ~$0.30/1000 chars | Premium ads, character voices |
| Service | Cost | Best For |
|---|---|---|
| Lyria 2 (Google) | $0.0001/sec | AI-generated custom music |
| Jamendo | Free (CC) | Pre-made royalty-free tracks |
| Model | Cost | Output |
|---|---|---|
| Robust Video Matting | $0.046/run (flat) | Alpha mask → ffmpeg merge |
| Bria Video | $0.14/sec | Native transparent WebM |
