46 MB

Ctrl+K

1 contributor

History: 12 commits

Celsia

Model card: throughput across CPU, MPS and CUDA

e0a4e3e verified 5 days ago

.claude
Upload folder using huggingface_hub 5 days ago
.github
Upload folder using huggingface_hub 5 days ago
.hypothesis
Complete CPU run: all 16 models on all three devices, and the throughput finding 5 days ago
CPU agree
Add CPU to the device comparison and a generated evaluation figure across CUDA, CPU and MPS 5 days ago
CPU bench
Complete CPU run: all 16 models on all three devices, and the throughput finding 5 days ago
CUDA agree
CUDA benchmarks on an RTX 5090: checkpoints, agreement runs, shared init weights, and the corrected comparison 5 days ago
CUDA bench
CUDA benchmarks on an RTX 5090: checkpoints, agreement runs, shared init weights, and the corrected comparison 5 days ago
MPS agree
CUDA benchmarks on an RTX 5090: checkpoints, agreement runs, shared init weights, and the corrected comparison 5 days ago
MPS bench
CUDA benchmarks on an RTX 5090: checkpoints, agreement runs, shared init weights, and the corrected comparison 5 days ago
benchmarks
Complete CPU run: all 16 models on all three devices, and the throughput finding 5 days ago
docs
Complete CPU run: all 16 models on all three devices, and the throughput finding 5 days ago
dossier
Add CPU to the device comparison and a generated evaluation figure across CUDA, CPU and MPS 5 days ago
examples
Upload folder using huggingface_hub 5 days ago
init weights
CUDA benchmarks on an RTX 5090: checkpoints, agreement runs, shared init weights, and the corrected comparison 5 days ago
scripts
CUDA benchmarks on an RTX 5090: checkpoints, agreement runs, shared init weights, and the corrected comparison 5 days ago
src
CUDA benchmarks on an RTX 5090: checkpoints, agreement runs, shared init weights, and the corrected comparison 5 days ago
tests
Complete CPU run: all 16 models on all three devices, and the throughput finding 5 days ago
viewer
Remove viewer/node_modules: a nested path my exclude pattern missed 5 days ago
.gitattributes

1.93 kB
Add CPU to the device comparison and a generated evaluation figure across CUDA, CPU and MPS 5 days ago
.gitignore

339 Bytes
Upload folder using huggingface_hub 5 days ago
AGREEMENT.md

3.82 kB
CUDA benchmarks on an RTX 5090: checkpoints, agreement runs, shared init weights, and the corrected comparison 5 days ago
BENCH-README.md

13.2 kB
Complete CPU run: all 16 models on all three devices, and the throughput finding 5 days ago
BENCHMARK-DESIGN.md

6.28 kB
Upload folder using huggingface_hub 5 days ago
BENCHMARKS.md

8.08 kB
Upload folder using huggingface_hub 5 days ago
CHANGELOG.md

33.4 kB
CUDA benchmarks on an RTX 5090: checkpoints, agreement runs, shared init weights, and the corrected comparison 5 days ago
COMPARISON.md

2.1 kB
CUDA benchmarks on an RTX 5090: checkpoints, agreement runs, shared init weights, and the corrected comparison 5 days ago
CONTRIBUTING.md

3.35 kB
Upload folder using huggingface_hub 5 days ago
DEBUG.md

41.6 kB
Upload folder using huggingface_hub 5 days ago
DESIGN.md

15.9 kB
Upload folder using huggingface_hub 5 days ago
LICENSE

11.4 kB
Upload folder using huggingface_hub 5 days ago
LTI.md

8.98 kB
Upload folder using huggingface_hub 5 days ago
NOTICE

8.15 kB
Upload folder using huggingface_hub 5 days ago
PLAN.md

54.4 kB
Upload folder using huggingface_hub 5 days ago
README.md

11.7 kB
Model card: throughput across CPU, MPS and CUDA 5 days ago
RESULTS.md

6.7 kB
Upload folder using huggingface_hub 5 days ago
SCORECARD.md

2.48 kB
Upload folder using huggingface_hub 5 days ago
VIEWER.md

9.07 kB
Upload folder using huggingface_hub 5 days ago
pyproject.toml

4.65 kB
Add CPU to the device comparison and a generated evaluation figure across CUDA, CPU and MPS 5 days ago