Data◆Article

Fable's AI system wrote the first genuine and fastest megakernel ever submitted to KernelBench-Mega, hitting an 18.71X speedup with just one cooperative kernel launch per decoded token, versus rivals that needed 4 to 14 separate launches.

Fable's AI-written CUDA megakernel achieved an 18.71x speedup on KernelBench-Mega using only one kernel launch per token, beating attempts by Claude Opus 4.8, GLM-5.2, and GPT-5.5, signaling progress toward AI automating its own R&D. ✦ AI generated

Jack Clark (Import AI) · Import AI · 2026-07-06 · original ↗

Export .md Export .txt

This solution is particularly impressive because "torch.profiler shows exactly ONE cooperative kernel launch per decoded token". By comparison, every other high-scoring entry decomposed the problem into anywhere from 4 to 14 separate kernel launches per token.

Read full article ↗excerpt · fair-use quotation

✳ CUDA ✳ Fable ✳ KernelBench-Mega

Around this claim

This moment responds to

extends → Fable's AI-written megakernel is the fastest and most efficient solution ever submitted to KernelBench-Mega, using a single cooperative kernel launch per token where every other high-scoring entry needed 4 to 14 — a sign AI is closing in on the R&D tasks that underlie recursive self-improvement.Jack Clark (Import AI) · Import AI