One control layer
for your AI.
Connect supported AI services and use their capabilities through one controlled platform and developer experience.
OpenAI-compatible API for supported GhostMind capabilities.
Three steps to unified AI
GhostMind sits between your application and your AI providers, normalizing access and control.
Connect
Connect supported AI accounts to your workspace. GhostMind manages credentials and health.
Control
GhostMind manages access, routing, and capability normalization across providers.
Build
Use the Playground or the API from your application. One interface, consistent behavior.
What you can do today
Current capabilities available through GhostMind, derived from your connected providers.
Chat
Stream responses from connected AI accounts with real-time SSE streaming.
Conversations
Maintain context across messages with conversation continuity and history.
Projects
Organize conversations with shared file context, instructions, and memory policies.
Files & Attachments
Upload files to conversations or projects for AI retrieval and context.
Generated Assets
Receive and download files generated by the AI during conversations.
Speech Generation
Generate audio from text with async jobs, multiple voices, and dialects.
Transcription
Convert audio to text with privacy-first processing — no transcript storage.
Usage & Control
Monitor API consumption with per-workspace usage summaries and records.
Connected AI services
GhostMind routes requests through your own connected provider accounts.
ChatGPT
Chat completions, conversations, projects with file context, file attachments, and audio transcription.
Google Savio
High-quality neural text-to-speech with Arabic dialects and multilingual voice support.
Build with a familiar API
OpenAI-compatible API for supported GhostMind capabilities. Use the OpenAI SDK with a custom base URL.
# Chat completion
curl https://ghostmind.optdmsa.com/v1/chat/completions \
-H "Authorization: Bearer sk-gm-..." \
-H "Content-Type: application/json" \
-d '{
"model": "auto",
"messages": [
{"role": "user", "content": "Hello!"}
]
}'Use the OpenAI SDK with supported GhostMind capabilities.
Real-time streaming responses with Server-Sent Events.
Maintain context across multiple requests.
Submit, poll, and download speech generation jobs.
Context that persists
Organize conversations into projects with shared file context. The AI can read and cite uploaded files from the first message.
- Upload files to a project for AI retrieval
- Conversations maintain context across messages
- File citations appear in AI responses
- Memory policies control how history is used
Speech and transcription
Generate audio from text and transcribe audio to text through one API.
Speech Generation
Convert text to audio with multiple voices and dialects. Async job lifecycle with polling and download.
Transcription
Upload audio and receive text. Privacy-first: no transcript storage, audio deleted after processing.
Governed by design
GhostMind centralizes provider access and operational control.
Workspace Isolation
Each workspace is fully isolated — connections, keys, and data never cross boundaries.
Scoped API Keys
API keys are workspace-scoped with configurable permissions for chat, conversations, and audio.
Encrypted Credentials
Provider credentials are encrypted at rest using Fernet symmetric encryption.
Audit Logging
Every mutating operation is recorded with audit logs for governance and compliance.
Role-Aware Access
Owner, admin, and member roles control who can manage connections, keys, and settings.
Private Assets
Generated assets are tenant-isolated with cross-tenant access denied and path traversal prevention.
Common questions
What is GhostMind?
GhostMind is an AI gateway that provides an OpenAI-compatible API for supported capabilities. It routes requests through your connected AI provider accounts with workspace isolation, usage metering, and quota enforcement.
Which AI services can I connect?
Currently, GhostMind supports ChatGPT (chat, conversations, projects, transcription) and Google Savio (speech generation). New providers can be added through the adapter interface.
Can I use it through an API?
Yes. GhostMind provides an OpenAI-compatible API. You can use the OpenAI SDK with a custom base URL, or make direct HTTP requests with any language.
Does GhostMind replace my AI provider?
No. GhostMind is a control layer that sits between your application and your AI provider accounts. You connect your own accounts — GhostMind manages access, routing, and capability normalization.
How do projects work?
Projects group conversations with shared file context. Upload files to a project, and the AI can read and cite them from the first message in any project conversation.
Can I generate speech?
Yes. With a connected Google Savio account, you can generate audio from text with multiple voices and dialects using async jobs.
How is access controlled?
Access is controlled through workspace isolation, scoped API keys, and role-based permissions (owner, admin, member). Provider credentials are encrypted at rest.