Posts tagged

From 479 preemptions to serving 128k contexts out of a 64 GB disk cache — including the upstream bug I had to patch to get there.

From 26 t/s to 33 t/s with a 2x context window — what actually moved the needle, and the tradeoffs that didn't show up in the benchmark.

Is WSL 2's Vmmem process consuming all your system RAM? Here's how to use the .wslconfig file to set hard limits and regain control of your Windows resources.