GPU unlock tested: ~50% TFLOPS boost on BF16/FP16/TF32 with identical outputs
johnowhitaker · x · 2026-10-26
Responding to unixpickle's skepticism about GPU "unlock" tweaks, johnowhitaker ran his own test: a 50% TFLOPS boost on BF16, FP16 and TF32 in his matmul benchmark, with identical outputs—suggesting the unlocks are legitimate performance releases, not broken numerics or misinformation.
Related event: GPU 'Unlock' Speedups Verified: ~50% Faster With Identical Outputs(2 posts)→
More from Infra
- RTX 4090 workstation vs AI subscriptions: an editor's break-even math — iamthecarguy1234 · 2026-10-26
- Cerebras CS-4 teardown: 2x bandwidth and power, 50% fewer parts vs CS-3 — jwt0625 · 2026-10-26
- theo doubts Windows ARM: porting 20 years of tech debt vs Intel improving — wateriscoding · 2026-10-26
- TokenSpeed's Three-Language Stack: C++ for Correctness, Rust for Agents, Python for Glue — zhyncs42 · 2026-10-26
- Speculative Decoding Explained: a 30% Wrong Guesser Never Changes the Output — anbarth · 2026-10-26
- You Can Train a 1B Model in 3 Weeks: Dev Shares 15.9-Day Training Run Log — arthurcolle · 2026-10-26