performance-improvement

GPT-Realtime-2 demonstrates a 15 percentage point improvement over version 1.5 on the Big Bench Audio benchmark, approaching saturation levels.

OpenAI released two new embedding models: text-embedding-3-small (5x cheaper than ada-002 with 40%+ MIRACL improvement) and text-embedding-3-large (best performance with up to 3072 dimensions). Both models show significant performance gains on standard benchmarks while reducing costs.