Issue 2026-11-17 · Industry · 开源 · 芯片
RunAnywhere open-sources TrnRT inference engine for AWS Trainium

RunAnywhere (YC W26) open-sourced TrnRT, an LLM inference engine for AWS Trainium. It serves two GLM-5.3 flash instances on a single trn2.48xlarge, processing 3.57 million tokens per minute on AgentX benchmarks. The team built it because AWS's existing stack lacked support for GLM models.
Read original ↗