Together AI Serverless Model Bring-Up: DeepSeek-V4-Pro, Gemma-4, GLM-5.1 Added

AI Impact Summary

Together AI has initiated a significant serverless model bring-up, adding several new models including deepseek-ai/DeepSeek-V4-Pro, google/gemma-4-31B-it, and zai-org/GLM-5.1. This expansion, coupled with the deprecation of Qwen/Qwen3-VL-8B-Instruct, Qwen/Qwen3-235B-A22B-Thinking, and mistralai/Mixtral-8x7B-Instruct-v0.1, represents a shift in the available serverless compute options. The introduction of Dedicated Container Inference (DCI) further expands deployment flexibility, allowing users to containerize and scale custom models.