Show HN: TrnRT – An open-source LLM inference engine for AWS Trainium
Hi HN, Sanchit here from RunAnywhere (YC W26). We just open-sourced TrnRT: Trainium RunTime, our inference engine for AWS Trainium. We started 3 weeks back, and currently we’re able to serve and run GLM 5.3 flash (FP8) on AWS trn2.48xlarge, and will bring more models support soon. It’s open to contributions :) What it does today: - Serves 2 instances of GLM-5.3 flash on a single box, 8 trn2 with a cache aware router. - Can process 3.57M TPM (tokens per minute) on SemiAnalysis's AgentX benchmarks
阅读原始来源来源记录
指标快照
points · 42 votes
2026-11-18 00:29:12.837Z
cumulative since publication
研究分析
项目记录
仓库:RunanywhereAI/trnrt。 来源描述:Hi HN, Sanchit here from RunAnywhere (YC W26). We just open-sourced TrnRT: Trainium RunTime, our inference engine for AWS Trainium. We started 3 weeks back, and currently we’re able to serve and run GLM 5.3 flash (FP8) on AWS trn2.48xlarge, and will bring more models support soon. It’s open to contributions :) What it does today: - Serves 2 instances of GLM-5.3 flash on a single box, 8 trn2 with a cache aware router. - Can process 3.57M TPM (tokens per minute) on SemiAnalysis's AgentX benchmarks
采用记录与比较对象
来源票数: 42 votes (2026年11月18日 UTC; cumulative since publication);前次同口径记录为 42(2026年11月18日 UTC),两个观测值不等于需求增长。 可供比较的近期同类记录:“alicebrown323/pro-tools-desktop” (Trendshift, 2026年11月18日);“jingyaogong/minimind” (Trendshift, 2026年11月18日)。主题相近仅用于导航,未核验其是否属于同一需求或独立证据。
采用判断
应与它要替代的工具或人工流程比较。现有记录支持发现项目,尚不能确认可靠性、运行成本优势或商业采用。
尚未核验的门槛
尚未核验:许可证权限、硬件要求、安装步骤、维护状态,以及包含失败样例的代表性任务。
下一步试验
选择一个真实任务和现有替代方案,记录 commit、环境、输入、输出、耗时与失败行为;保留结果后再给出采用建议。