https://openrelay.tech/blog Deep dives into distributed GPU architecture, engineering decisions, and the technology powering OpenRelay's fault-tolerant inference platform. en-us Wed, 21 Oct 2026 14:42:35 GMT https://openrelay.tech/blog/gpu-layers-explained https://openrelay.tech/blog/gpu-layers-explained What llama.cpp's --n-gpu-layers (-ngl) puts on the GPU, how to size it from free VRAM and the GGUF file, and what fails when it is set too high. Sun, 04 Oct 2026 00:00:00 GMT llama.cpp https://openrelay.tech/blog/gemma-4-agentic-kernel-engineering https://openrelay.tech/blog/gemma-4-agentic-kernel-engineering How OpenRelay's kernel agents got 1.5x more traffic out of each AMD MI355X serving Gemma 4 31B, at the same latency target and the same accuracy. We are open-sourcing all of it. Tue, 29 Sep 2026 00:00:00 GMT Kernel Engineering https://openrelay.tech/blog/openrelay-cli-0-4 https://openrelay.tech/blog/openrelay-cli-0-4 orl 0.4 adds GPU pods, VM sizing checks before launch, per-VM burn rates, and expedited batch inference. Install it and rent a GPU from your terminal. Sat, 26 Sep 2026 00:00:00 GMT Release https://openrelay.tech/blog/openrelay-backed-by-y-combinator https://openrelay.tech/blog/openrelay-backed-by-y-combinator We're building the CDN of inference: a distributed GPU network that makes fast, affordable, fault-tolerant AI compute available to everyone. Here's why, and what comes next. Sat, 06 Jun 2026 00:00:00 GMT Company