New
2 New LLMs
Kimi K3
, Moonshot AI's most capable model is now available in the model selector for chats, agents, and workflows. Its responses take a little longer to arrive, but extra thinking behind them delivers stronger, more thorough output. "
Kimi K3 is a 2.8T-parameter model built on our Kimi Delta Attention and Attention Residuals, with native vision capabilities and a 1-million-token context window. It is the world's first open 3T-class model, designed for frontier intelligence across long-horizon coding, knowledge work, and reasoning."
– Moonshot AI
GLM 5.2 Fast
(ZAI) is now available in the model selector for chats, apps, agents, and workflows. A speed-optimized serving tier of GLM-5.2, hosted by Baseten and built for real-time, latency-sensitive agent workloads. "
GLM-5.2 Fast: our GLM-5.2 Model API designed for the most demanding real-time use cases. The Fast tier delivers 2-3x higher TPS than the standard GLM-5.2 MAPI.
" – Baseten