Blue Horizon AI
A Persian-first platform that puts the world's leading AI models behind one interface and one wallet.
The problem it solved
Iranian users who want today's frontier AI models face three walls: no international payment card, a separate subscription for every tool, and English-only interfaces. The platform removes all three with one Persian interface and one toman-funded wallet. For the operator it guarantees that concurrent requests can never spend beyond a user's balance and that a model sold as free never costs the user anything.
Similar problems it can solve
The same architecture fits any organisation that wants to hand AI models to staff or customers under cost control: an internal LLM gateway with per-team quotas and budgets, wholesale resale of AI access through partners, or a localised AI service for a market cut off from global payment rails.
Blue Horizon AI brings dozens of language, image, video and audio models under one Persian interface, with toman payment and a single USD-denominated wallet. Next to chat and creative studios it exposes an OpenAI-compatible API, so developers connect by changing a base URL. Its standout strength is the money engine: every request is held before it runs and settled after, and a built-in auditor re-checks the books every six hours.
- Multi-provider model gateway behind one catalogue; free models appear under Persian names in the «افق آبی» tier.
- USD wallet with atomic hold-then-settle billing, a reseller layer and a self-auditing financial engine.
- .NET 10 Razor Pages web, a native Android app in Jetpack Compose, Telegram bots and an OpenAI-compatible /v1 API.
The product
Blue Horizon AI is a Persian-first AI services platform. One account, one wallet and one Persian interface give access to dozens of language, image, video, music and speech models, paid in toman.
One platform, many components
- Chat & assistants: encrypted conversation history, private mode, file attachments, an assistant marketplace and agent chains
- Studios: image, video and music generation, plus a multi-scene Film studio, a Paper studio where a team of models drafts, reviews and revises a citable paper, and a Podcast studio
- Developer API: an OpenAI-compatible
/v1(chat, embeddings, media) with scoped personal keys - Android client: native Kotlin / Jetpack Compose app over a dedicated mobile API
- Operations: admin console, reseller layer, Telegram bots, community and blog
Engineering depth
- A multi-provider gateway merges several upstream catalogues into one model list and routes each request to the right backend; provider identity never reaches a customer surface.
- A USD-micro wallet reserves each request's worst-case cost atomically before it runs and settles the real cost after, so parallel requests cannot overspend.
- A financial invariant auditor re-verifies live rows every six hours (free models charged nothing, no negative wallets, holds balance) and pages on any violation.
- Privacy by design: AES-256-GCM encryption at rest with a per-conversation key, so deleting a conversation crypto-shreds it.
Technology
.NET 10 · ASP.NET Core Razor Pages · EF Core · SQL Server · Kotlin / Jetpack Compose · OpenAI-compatible API · Azure Key Vault