Skip to content

[Docs] RoPE gets its own Benchmarks page, after Elementwise #44

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

Merged
lcy-seso merged 1 commit into from
Sep 4, 2026
Merged

[Docs] RoPE gets its own Benchmarks page, after Elementwise #44

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
6 changes: 3 additions & 3 deletions hooks.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -78,9 +78,9 @@ def on_page_markdown(markdown, page, config, files):
# page reaches the nav without editing this list.
_BENCH_ORDER = [
"index.md", "reading.md",
"elementwise.md", "reduction.md", "normalization.md", "conv-pool.md",
"gemm.md", "quantization.md", "attention.md", "moe.md",
"linear-attention.md", "ssm.md", "other.md",
"elementwise.md", "rope.md", "reduction.md", "normalization.md",
"conv-pool.md", "gemm.md", "quantization.md", "attention.md",
"moe.md", "linear-attention.md", "ssm.md", "other.md",
]


Expand Down
15 changes: 8 additions & 7 deletions scripts/gen_bench_pages.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -95,7 +95,7 @@ def tier_of(tag: str) -> str:
"ssm": "SSM", "scan": "Scan", "normalization": "Normalization",
"moe": "MoE", "linear_algebra": "GEMM", "reduction": "Reduction",
"elementwise": "Elementwise", "convolution": "Convolution", "pool": "Pooling",
"quantization": "Quantization", "positional": "Positional Encoding",
"quantization": "Quantization", "positional": "RoPE",
"fft": "FFT", "mhc": "MHC", "topk": "Top-k", "other": "Other",
}
# Ops a family lists in a fixed order rather than by verdict, named as the page
Expand All @@ -108,12 +108,14 @@ def tier_of(tag: str) -> str:
"BmmFp8NKFwd", "BmmFp8KNFwd"],
}
# (slug, page title, families in display order). One page per family, ordered
# from the simplest op to the most composed: pointwise, then reductions over an
# axis, then the normalizations built on them, then the sliding-window ops, then
# the matrix-multiply pages, then the sequence-mixing ops. `Other` is last and
# holds every family too small to carry a page.
# from the simplest op to the most composed: pointwise, then the positional
# rotation applied to one tensor, then reductions over an axis, then the
# normalizations built on them, then the sliding-window ops, then the
# matrix-multiply pages, then the sequence-mixing ops. `Other` is last and holds
# every family too small to carry a page.
DATA_PAGES = [
("elementwise", "Elementwise", ["elementwise"]),
("rope", "RoPE", ["positional"]),
("reduction", "Reduction", ["reduction"]),
("normalization", "Normalization", ["normalization"]),
("conv-pool", "Conv & Pool", ["convolution", "pool"]),
Expand All @@ -123,8 +125,7 @@ def tier_of(tag: str) -> str:
("moe", "MoE", ["moe"]),
("linear-attention", "Linear Attention", ["linear_attention"]),
("ssm", "SSM", ["ssm"]),
("other", "Other",
["positional", "fft", "mhc", "topk", "scan", "other"]),
("other", "Other", ["fft", "mhc", "topk", "scan", "other"]),
]
_KEYWORD_FAMILY = [
(("mamba", "ssd", "ssm"), "ssm"),
Expand Down
Loading