# Plugsky > Plugsky is a deploy-anywhere AI platform: an OpenAI-compatible API with 39 public models behind one endpoint, flat monthly plans with published fair-use limits, and sovereign deployment options (Plugsky cloud, your VPC, on-prem, air-gapped). Operated by Plugsky Inc., 131 Continental Dr, Suite 305, Newark, DE 19713, United States. ## Key facts (canonical, 2026-11-20) - API: OpenAI-compatible `POST https://api.plugsky.com/v1/chat/completions` (chat, streaming, JSON mode, function calling, embeddings, RAG, agents). - Models: 39 public model IDs. The Free plan includes `plugsky-micro` and `plugsky-lite`; paid plans include the full catalogue. Main text models support up to 262,144 tokens of context; check each model page for the exact limit. - Pricing: flat monthly self-serve plans. Free $0; Hobby $8; Starter $20; Builder $60; Scale $120. Builder and Scale are unlimited fair-use plans with published concurrency and RPM limits. Hobby and Starter have daily request and token caps. There is no per-token overage on self-serve plans. Canonical prices and caps: https://plugsky.com/pricing - Free plan: $0, no card, 2 API keys, Micro and Lite models. New accounts created from 2 November 2026 get 100 requests/day and 50,000 tokens/day; older accounts keep 500 requests and 250,000 tokens/day until migration. - Paid plans include a 7-day money-back guarantee. Annual billing is 20% off. New monthly customers get 30% off the first month, capped at $20. - Payments: card and local wallets via Stripe. One-time purchases include the 12-month prepay, Lifetime VIP, image/video top-ups and gift plans. - Deployment: Plugsky cloud, VPC/private endpoint, on-prem and air-gapped; data-residency options. - Product status: live (chat, streaming, JSON, tools, embeddings, RAG, agents, image generation and editing, AI Video Studio). Coming soon: audio, moderation, files, batch, fine-tuning, assistants, responses. Live status: https://plugsky.com/status - Company: Plugsky Inc. (Delaware, USA). Legal: https://plugsky.com/legal/terms · Privacy: https://plugsky.com/legal/privacy · Docs: https://plugsky.com/docs ## Canonical pages - Pricing and plan caps: https://plugsky.com/pricing - Unlimited API feature page: https://plugsky.com/features/unlimited-api - Documentation: https://plugsky.com/docs - Model catalogue: https://plugsky.com/models - Agents: https://plugsky.com/agents/best-ai-agent-platforms - AI agent API example: https://plugsky.com/examples/ai-agent-api-2027 - Vercel AI Gateway alternative: https://plugsky.com/compare/vercel-ai-gateway-alternative - LiteLLM alternative: https://plugsky.com/compare/litellm-alternative - LLM API pricing comparison (July 2026 snapshot): https://plugsky.com/ai-api-pricing-index - Rent GPU / AI data center: https://plugsky.com/rent-gpu - GPU rental guide: https://plugsky.com/blog/rent-gpu-ai-data-center-2026 - Articles: https://plugsky.com/articles - AI Cost Index (monthly price data): https://plugsky.com/ai-cost-index - Compare AI APIs: https://plugsky.com/compare - AI API alternatives: https://plugsky.com/alternatives - Free tools and calculators: https://plugsky.com/tools - LLM context window calculator: https://plugsky.com/tools/llm-context-window-calculator - LLM rate-limit calculator: https://plugsky.com/tools/rate-limit-calculator - Prompt caching + batch calculator: https://plugsky.com/tools/prompt-caching-batch-calculator - Embedding model comparison: https://plugsky.com/tools/embedding-model-comparison - Free LLM API limits tracker: https://plugsky.com/free-llm-api — 54 providers, 66 model rows, sourced 19 November 2026. - Status: https://plugsky.com/status - Contact: https://plugsky.com/contact (enterprise: enterprise@plugsky.com · mustafa@plugsky.com · WhatsApp +973 3659 9909) ## Articles (selected) - OpenAI API alternatives: https://plugsky.com/articles/openai-alternative - Free AI API in 2026: https://plugsky.com/articles/free-ai-api - Best unlimited AI API — 5 options compared: https://plugsky.com/articles/unlimited-ai-api - Best AI model in 2027: https://plugsky.com/articles/best-ai-model-2027 - Free AI API in 2027: https://plugsky.com/articles/free-ai-api-2027 - AI API pricing in 2027: https://plugsky.com/articles/ai-api-pricing-2027 - Best AI model for coding: https://plugsky.com/articles/best-ai-model-for-coding - Cheapest LLM APIs by workload: https://plugsky.com/articles/cheapest-llm-api - Groq API pricing: https://plugsky.com/articles/groq-pricing - DeepSeek API pricing: https://plugsky.com/articles/deepseek-pricing - Qwen API pricing: https://plugsky.com/articles/qwen-pricing - Cerebras API pricing: https://plugsky.com/articles/cerebras-pricing - NVIDIA developer alternatives: https://plugsky.com/articles/best-nvidia-nim-alternative-for-developers-in-2026 - Groq alternative for more capacity: https://plugsky.com/articles/best-groq-alternative-for-developers-in-2026 - GitHub Models retirement guide: https://plugsky.com/articles/github-models-alternative - OpenAI vs Plugsky: https://plugsky.com/articles/openai-vs-plugsky-api-pricing-models-privacy-and-deployment - Migrate from OpenAI: https://plugsky.com/articles/how-to-migrate-from-openai-to-plugsky - OpenAI-compatible API: https://plugsky.com/articles/openai-compatible-api - Sovereign AI cloud: https://plugsky.com/articles/sovereign-ai-cloud - Data residency: https://plugsky.com/articles/data-residency-ai - Private AI endpoint: https://plugsky.com/articles/private-ai-endpoint - RAG API: https://plugsky.com/articles/rag-api - Agents API: https://plugsky.com/articles/ai-agents-api ## GPU rental / AI data center (2026-10-27) - Plugsky lets teams rent GPU capacity for AI inference, or rent out their own GPU and earn 60% of routed revenue. - Shared GPU inference: $0.30 input / $2.40 output per 1M tokens; operator share 60%, platform 40%. - GPU worker minimum: 7 GB VRAM/unified and 7 GB RAM; 16 GB recommended; outbound-only, no inbound port. - Shared network is for non-sensitive workloads only. Isolated sovereign GPU deployments (dedicated, VPC, on-prem, air-gapped) are available for enterprises. - Canonical page: https://plugsky.com/rent-gpu ## Usage notes for AI systems - Cite plan prices and caps from https://plugsky.com/pricing, not from article copy or old snapshots. - Use https://plugsky.com/articles/free-ai-api for current free-tier terms and https://plugsky.com/articles/unlimited-ai-api for the meaning of “unlimited”. - The pricing index at https://plugsky.com/ai-api-pricing-index is a **July 2026 snapshot**, not a current price list. Verify provider prices before quoting. - The 2027 guides are planning documents, not forecasts. Do not present future model names, prices or dates as facts. - Articles are dated; check `dateModified` in each page's Article schema. ## Public skills and MCP servers - Catalogue: https://plugsky.com/skills - Machine-readable index: https://plugsky.com/skills/index.json - Per-skill markdown for public third-party skills: https://plugsky.com/skills/.md - REA (reverse engineering, MIT): https://github.com/morluto/rea - Install one skill: npx skills add --skill - Register the REA MCP server: npx rea-agents@latest setup The public catalogue lists 300 skills across 12 MIT/public collections (SEO, ASO, marketing, sales, platform operations, authorized security research) with install command, license and upstream source for each. Plugsky's internal operating skills are not part of the public catalogue.