Together AI
ThunderAgent
ThunderAgent is a program-aware scheduling system designed to optimize high-throughput agentic inference. By treating agentic workflows as schedulable programs rather than independent requests, it effectively eliminates KV cache thrashing and improves multi-node scaling.
DSGym v1.0
DSGym is a comprehensive framework designed to evaluate and train data science agents within self-contained execution environments. It offers a modular architecture that facilitates the addition of tasks, agent scaffolds, and tools, positioning it as a live, extensible testbed for data science agents.
Together AI Voice Finder vinitial release
A developer tool to search, filter, and audition over 600 text-to-speech voices across leading TTS models using natural-language prompts or audio samples, making it easy to find the perfect voice for any application.