Your conversations.
Your models.
Your rules.
One interface for OpenAI, Anthropic, Google, Mistral, Ollama, and any OpenAI-compatible endpoint. Free to start. BYOK supported.
Everything you need. Nothing you don’t.
Every Model, One Conversation
Switch between OpenAI, Anthropic, Google, Mistral, and Ollama mid-conversation. Use custom endpoints for anything OpenAI-compatible.
Agents You Can Build
Create custom assistants that research, write, code, and collaborate — each with their own tools and memory.
Code That Runs Itself
Agents execute Python, browse the web, and produce results — while you watch every step in real time.
Your Instance. Your Data.
Self-hosted or cloud, your conversations stay yours. No training, no telemetry, no surprises.
Your keys. Your models. Your API costs.
Use Sapient with your own API keys from any provider. You pay the provider directly — no markup, no middleman, no surprises on your bill.
Start free. Scale when you’re ready.
Free Tier
- Access to built-in models
- Limited daily message quota
- Custom agents
- Code execution
- BYOK support
- Synapse quota sharing
Synapse + Sapient
- Everything in Free
- Synapse quota applies to Sapient
- Higher message limits
- Priority model access
- Advanced agent features
- BYOK + unlimited custom endpoints
What people are building with it.
Draft anything
From emails to essays. Get a first draft in seconds, then make it yours.
Understand faster
Summarize papers, analyze data, explore ideas. The thinking happens with you.
Build with intelligence
Debug, refactor, prototype. Agents that understand your codebase and ship.
Make sense of data
Upload documents, extract insights, compare sources. The analysis runs in real time.
Work together
Share conversations, build shared agents, collaborate on projects across your team.
Keep it yours
Sensitive conversations, proprietary code, personal thoughts. Nothing leaves your control.
Ready when you are.
An open-source chat platform that respects your privacy, supports every model, and runs on your terms.