KernelBench hard · H100
KDA CUTLASS GLM-5.2
slowdid not score
harnesszai-claudeagent session45mtotal wall45mcheck—benchmark—output tokens—regimecompute
Per-shape vs governing ceilingeach shape graded against whichever binds — bf16 compute or HBM bandwidth
No per-shape benchmark data archived for this run.
20260618_170834_zai-claude_glm-5.2_02_kda_cutlass