> Canonical page: https://posttrainllm.com/speedup

[← posttrainllm](https://posttrainllm.com/) · [roadmap](https://posttrainllm.com/roadmap) · [devlog](https://posttrainllm.com/devlog) · [docs](https://posttrainllm.com/docs/)

# A verified Large-preset win — and the historical curve around it

 A frozen ABBA run on an Apple M5 Pro measured 10.67×
 WebGPU-over-WASM speedup on Large, with 4.72% maximum paired final-loss
 drift and zero runtime errors. That point is verified for this machine and
 configuration. The older Small, Medium, and XL values remain historical.

Small · d=96 WASM SIMD (mt)

2.6× historical

Medium · d=128 WASM SIMD (mt)

6.8× historical

Large · d=192 WASM SIMD (mt)

10.67× measured · M5 Pro

XL · d=256 WASM SIMD (mt)

12.1× historical

 10.67×
 measured at Large: 1462.5 ms/step WASM versus 137.1 ms/step WebGPU medians.

 Current source: evals/verified-wins/webgpu-paired-result-v1.json .
 Chrome 151 identified apple · metal-3 , rejected software and
 fallback adapters, and ran WASM → WebGPU → WebGPU → WASM on identical inputs.
 The retained 2.6×–12.1× multi-preset curve is still historical evidence only.

## The hypothesis encoded by the curve

Every training step costs fixed_overhead + math(model_size) . Larger matmuls should amortize GPU dispatch overhead better. The historical curve is consistent with that mechanism, but the mechanism does not verify the retained numbers.

scalar WASM baseline

1.0×

measured

+ WASM SIMD

1.6×

measured

+ multi-thread (4 workers)

3.2×

measured

+ WebGPU full stack — Small (d=96)

~8.3×

historical · unqualified

+ WebGPU full stack — Medium (d=128)

~22×

historical · unqualified

+ WebGPU full stack — Large (d=192)

~30×

historical · unqualified

+ WebGPU full stack — XL (d=256)

~39×

historical · unqualified

Mega / Behemoth (projected)

≥ 15×

projected · blocked by Memory64 ABI bug

 The shipped default WebGPU path uses blocked4 matmul, vec4 loads,
 subgroup-cooperative reductions, and Flash Attention 2 forward and
 backward. Use the playground's live benchmark for exploration; use a
 paired JSON receipt before publishing a performance result.

 Full benchmark log + the three things that didn't work →
 [/devlog](https://posttrainllm.com/devlog) · [/roadmap](https://posttrainllm.com/roadmap)