Updates
Changelog
Follow our journey as we build the most efficient AI gateway.
June 2025v1.0 (Current)
Gateway & Playground Live
- Public no-signup playground launched
- High-performance Go gateway deployed
- 8 compression rules active (LLMLingua-2 integration)
- DistilBERT complexity router active
- Python and TypeScript SDKs published
- Marketing website & documentation hub live
May 2025v0.9 (Beta)
Core Compression Engine
- Initial Python compression service built
- Quality scorer model fine-tuned
- LlamaIndex & LangChain callbacks implemented
April 2025v0.1 (Alpha)
Proof of Concept
- Architecture design complete
- First successful compressed Claude-3 call