<!-- agent-source-envelope -->
# Artifact: Browser WebGPU Training Speedup | PTLLM

> A frozen ABBA run on an Apple M5 Pro measured a 10.67x Large-preset WebGPU speedup over WASM with 4.72% paired final-loss drift and zero runtime errors.

Canonical: https://posttrainllm.com/artifacts/browser-webgpu-speedup
<!-- /agent-source-envelope -->

Browser performance artifact

Report artifact

2026-09-02

The browser playground now has a retained, adapter-qualified performance win for one configuration. Two WASM and two WebGPU runs used the same Large preset, corpus, seed, and 20-step budget. This qualifies Large on the measured M5 Pro; the older Small/Medium/XL curve remains historical until separately reproduced.

WebGPU, WASM, browser, performance

## Headline Numbers

Evidence class applies to each value: measured = retained evaluation or
runtime result; historical = recorded result not freshly reproduced;
derived = calculated or decided from records; observed = current artifact
state; not-measured = an explicit evidence gap.

**WebGPU speedup:** 10.67x

measured · Large · Apple M5 Pro

Evidence: measured

**Loss drift:** 4.72%

maximum across two paired comparisons

Evidence: measured

**Hardware gate:** pass

Apple · metal-3 · no fallback

Evidence: measured

## Competitive Context

| System | Metric | Score | Size / Class | Comparable? | Readout |
| --- | --- | --- | --- | --- | --- |
| posttrainllm WebGPU | training step speedup | 10.67x | Large · d_model=192 · 20 steps | Direct | Median 137.1 ms/step across two WebGPU runs on the same Apple hardware and frozen inputs. |
| posttrainllm WASM SIMD | training step speedup | 1.0x | same browser model/config | Direct | Median 1462.5 ms/step across the two matching WASM runs. |
| Native Mac runtimes | browser training benchmark | not measured | MLX/Metal class | Not comparable | Native runtimes are the right competition for production throughput, but not for the browser-learning artifact. |

Direct rows share this artifact's eval setup. Directional rows are
useful market context but should not be read as leaderboard claims.

## Performance readout

| Variant | Result | Interpretation |
| --- | --- | --- |
| WASM · Large | 1462.5 ms/step median | Two matching runs |
| WebGPU · Large | 137.1 ms/step median | 10.67x measured speedup |
| Final-loss drift | 4.72% maximum | Passes the frozen <5% gate |
| Other presets | 2.6x–12.1x historical | Still unqualified |

## Release Blockers

## Evidence

- [Performance journey](https://posttrainllm.com/roadmap)

- [Performance docs](https://posttrainllm.com/docs/performance/performance)

- [Playground](https://posttrainllm.com/playground)

- [Paired result JSON](https://github.com/PostTrainLLM/posttrainllm/blob/main/evals/verified-wins/webgpu-paired-result-v1.json)

## Next Release Action

Keep Large as the measured M5 Pro win; reproduce Small, Medium, and XL independently before promoting the historical curve.
