October 2025
New Features
- 🚀 Added DeepSeek R1 reasoning model
- 📊 New batch API
- 🔧 Improved function calling support
Improvements
- ⚡ 30% faster response times for Llama 3.1 models
- 💰 Reduced pricing across all models
- 📈 Enhanced monitoring dashboard
- 🔐 Added SAML SSO for enterprise customers
Bug Fixes
- Fixed streaming issues with vision models
- Improved error messages for rate limits
- Fixed token counting for certain models
September 2025
New Features
- 🎯 Dedicated inference instances now available
- 🎨 Added Stable Diffusion 3 support
- 📝 Structured outputs with JSON schema validation
- 🌐 New data center in EU region
Improvements
- 🚀 50% faster cold starts
- 📊 Enhanced usage analytics
- 🔄 Auto-retry for failed requests
August 2025
New Features
- 🤖 Launched CheapestInference platform
- 💬 Chat completions API
- 🎯 Support for 50+ models
- 🔑 API key management
- 📚 Complete documentation
View all releases →