Skip to content

Align release evidence and publication hygiene #5

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
clocksmith wants to merge 1 commit into
base: main
Choose a base branch
from
Draft

Align release evidence and publication hygiene #5

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 4 additions & 4 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,13 +24,13 @@ npx doppler-gpu

## Evidence

Doppler has lower steady-state inference latency than Transformers.js in each
Doppler has higher steady-state inference throughput than Transformers.js in each
of the comparable browser WebGPU results indexed below.

![Metal and Vulkan browser WebGPU latency distributions](https://raw.githubusercontent.com/clocksmith/doppler/main/assets/doppler-webgpu-evidence.svg)
![Metal and Vulkan browser WebGPU throughput](https://raw.githubusercontent.com/clocksmith/doppler/main/assets/doppler-webgpu-evidence.svg)

For text, faster means lower decode ms/token; for retrieval it means lower
ms/embedding or ms/rerank. Every comparison passes its declared workload
Text results use decode `tok/s`; retrieval uses `embedding/s` or `reranks/s`.
Higher is faster. Every comparison passes its declared workload
correctness gate. Model loading is separate: Transformers.js loads the Vulkan
embedding and reranker artifacts faster. The
[competition scoreboard](https://github.com/clocksmith/doppler/blob/main/docs/model-competition-scoreboard.md) links every
Expand Down
158 changes: 106 additions & 52 deletions assets/doppler-webgpu-evidence.svg
View file
Open in desktop
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
8 changes: 4 additions & 4 deletions benchmarks/vendors/fixtures/g3-1b-p064-d064-t0-k1.compare.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -149,8 +149,8 @@
"dopplerDtype": null,
"dopplerModelSource": {
"source": "explicit",
"locator": "file:///Users/xyz/deco/doppler/models/local/gemma-3-1b-it-q4k-ehf16-af32",
"modelUrl": "file:///Users/xyz/deco/doppler/models/local/gemma-3-1b-it-q4k-ehf16-af32",
"locator": "file:///workspace/deco/doppler/models/local/gemma-3-1b-it-q4k-ehf16-af32",
"modelUrl": "file:///workspace/deco/doppler/models/local/gemma-3-1b-it-q4k-ehf16-af32",
"modelBaseDir": null,
"manifestSource": "models/local/gemma-3-1b-it-q4k-ehf16-af32/manifest.json",
"manifestSha256": "fcb3136e4188eec38e65f977359957bdc5f1b3590d63fba935b21474dfed0919",
Expand Down Expand Up @@ -358,7 +358,7 @@
"trainingBenchSteps": null,
"checkpointEvery": null,
"workloadType": null,
"modelUrl": "file:///Users/xyz/deco/doppler/models/local/gemma-3-1b-it-q4k-ehf16-af32",
"modelUrl": "file:///workspace/deco/doppler/models/local/gemma-3-1b-it-q4k-ehf16-af32",
"cacheMode": "warm",
"loadMode": "opfs",
"runtimeProfile": null,
Expand Down Expand Up @@ -1471,7 +1471,7 @@
"trainingBenchSteps": null,
"checkpointEvery": null,
"workloadType": null,
"modelUrl": "file:///Users/xyz/deco/doppler/models/local/gemma-3-1b-it-q4k-ehf16-af32",
"modelUrl": "file:///workspace/deco/doppler/models/local/gemma-3-1b-it-q4k-ehf16-af32",
"cacheMode": "warm",
"loadMode": "opfs",
"runtimeProfile": null,
Expand Down
8 changes: 4 additions & 4 deletions benchmarks/vendors/fixtures/qwen3-5-0-8b-p064-d064-t0-k1.compare.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1210,7 +1210,7 @@
}
},
"benchmarkPolicy": {
"source": "/Users/xyz/deco/doppler/benchmarks/vendors/benchmark-policy.json",
"source": "/workspace/deco/doppler/benchmarks/vendors/benchmark-policy.json",
"schemaVersion": 1,
"updated": "2026-03-29"
},
Expand Down Expand Up @@ -2194,7 +2194,7 @@
}
},
"benchmarkPolicy": {
"source": "/Users/xyz/deco/doppler/benchmarks/vendors/benchmark-policy.json",
"source": "/workspace/deco/doppler/benchmarks/vendors/benchmark-policy.json",
"schemaVersion": 1,
"updated": "2026-03-29"
},
Expand Down Expand Up @@ -3179,7 +3179,7 @@
}
},
"benchmarkPolicy": {
"source": "/Users/xyz/deco/doppler/benchmarks/vendors/benchmark-policy.json",
"source": "/workspace/deco/doppler/benchmarks/vendors/benchmark-policy.json",
"schemaVersion": 1,
"updated": "2026-03-29"
},
Expand Down Expand Up @@ -4163,7 +4163,7 @@
}
},
"benchmarkPolicy": {
"source": "/Users/xyz/deco/doppler/benchmarks/vendors/benchmark-policy.json",
"source": "/workspace/deco/doppler/benchmarks/vendors/benchmark-policy.json",
"schemaVersion": 1,
"updated": "2026-03-29"
},
Expand Down
Loading