Jectora

Jectora is an OpenAI-compatible AI gateway built for organizations. One API (https://api.jectora.com/openai/v1) routes requests across models from multiple providers with smart routing, hardware-aware inference optimization, and intelligent caching for lower cost and latency. Keep the official OpenAI JavaScript or Python SDK: set the Jectora base URL and API key, then choose a model by ID. Chat Completions, Responses, Embeddings, and model discovery are supported; capabilities vary by model. Organization workspaces separate API keys, balance, usage, and billing. Scope each key by model, spend, expiration, or source IP. The Dashboard shows spend, requests, and tokens. Pricing is pay as you go: $0 monthly fee, no minimum spend, prepaid credits with no fixed expiration while active, and no subscription, platform fee, or long-term commitment. Unit prices are in the model catalog (USD per 1M tokens for token-based models). Sign in with GitHub. Contact Contact by email.