deAPI

deAPI provides one API key and one REST endpoint to access 100+ open-source AI models for image generation, video creation, text-to-speech, audio and video transcription, and embeddings. The page states requests run on a decentralized GPU cloud, that new users receive $5 in free credits, that no credit card is required, and that pricing is usage-based with examples such as Text-to-Image from $0.00119 per image (512x512), Text-to-Speech at $0.77 per 1M characters, and Video-to-Text at $0.021 per hour. The page lists API docs, a Playground, SDK-style integrations with Python, TypeScript, and other frameworks, and MCP and n8n integrations. It also mentions rate limits (up to 300 requests per minute after payment), file limits for transcription (50MB video, 20MB audio), and that custom model upload is on the roadmap.

Key features

• Unified REST API with a single API key for multiple AI models
• 100+ open-source models for image, video, voice, music, and transcription
• Text-to-image generation, including FLUX and Z-Image models
• Text-to-speech in 20+ languages with multiple voice styles
• Video and audio transcription, including direct URLs from YouTube, X, TikTok, Twitch, and Kick
• Image-to-video generation
• Text-to-embedding generation with BGE M3
• Background removal, image upscaling, and OCR models
• Playground for testing without code
• Free credits ($5) for new users with no credit card required
• MCP integration and n8n support
• Usage-based pricing

Best suited for

• Developers building AI features into apps without managing infrastructure
• Startups and SaaS companies integrating generative AI
• Mobile app developers and Discord bot developers
• Chatbot and voice assistant developers
• Research teams and RAG systems that need transcripts and embeddings
• Teams transcribing YouTube or other video content for knowledge bases

Pricing

Usage-based. Free tier with $5 in free credits for new users; no subscriptions and no credit card required. Example rates on the page: Text-to-Image from $0.00119 per image (512x512), Text-to-Speech $0.77 per 1M characters, Video-to-Text $0.021 per hour. Enterprise pricing is available by contacting sales.