benchmark
01Command Claude Code
Pipe output to a file (e.g. bench.log in repo root) — /log-experiment will attach it.
7 tagged gpu, measured the same way as everything else here.
Command Claude Code
Pipe output to a file (e.g. bench.log in repo root) — /log-experiment will attach it.
Command Claude Code
Log the most recent experiment. Never skip — failures are as valuable as wins.
Command Claude Code
Iteratively improve solution/triton/sparsefused.py. Rules in CLAUDE.md are non-negotiable.
Command Claude Code
Launch the unattended / overnight AMK autoresearch driver on $ARGUMENTS (model [gpu] [minutes|iters]).
Command Claude Code
One-shot compile + verify a model into a CUDA megakernel via amk compile on $ARGUMENTS (model [gpu]).
Command Claude Code
Drive an interactive AMK propose -> eval -> keep/revert megakernel schedule session on $ARGUMENTS (model [gpu]).
Command Claude Code
Build a reference for the implementation details of this project. Use provided summarize tool to get summary of the files. Avoid reading the content of many files yourself, as we might hit usage limits. Do read the content of important files though. Use the returned summaries to create reference files in /ref…