Repository navigation
Documentation site on GitHub Pages (1bit-monster.github.io/engine) - #31
Merged
Merged
Conversation
…age each https://1bit-monster.github.io/engine/, built by .github/workflows/pages.yml from the repository's own README.md and docs/*.md, so it never drifts from them. - tools/site.py renders each doc (python-markdown: tables, fenced code, heading anchors) into site/template.html: a top bar, a sidebar in four groups (Start, Devices, Components, Project; any doc it does not place lands under "More", so none goes missing), previous / next, and a link to the page's source. Doc links become page links (a link labelled with a file name gets the page's name), other repository paths link to GitHub, 2-space nested lists nest. - site/style.css: system fonts, one teal accent, a 760px text column, styled tables and code, dark mode from the system setting; on phones the sidebar folds into a Contents toggle above the page. - pull requests only build the site; main deploys it. - tools/copyright.py: notice styles for .html and .css. - README links the site. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
… no native marker on the phone Contents toggle Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
bong-water-water-bong
added a commit
that referenced
this pull request
Sep 28, 2026
…208) * Laya scorer on HRX: ggmlc's laya on AMD's backend instead of Vulkan The router's GGUF scorer (#206) now runs on HRX0 (HRX + LOOM), not Vulkan: - third_party/ggmlc -> our fork 1bit-MONSTER/ggmlc 1bit/main (df08d1b0): an HRX build against a llama.cpp ggml, and a uid on every graph so HRX's graph-program cache and graph replay hit. - third_party/llama.cpp -> 22ceda3a (fork #31): the LOOM kernels ModernBERT needs on HRX (LayerNorm, strided/broadcast arithmetic, clamp, F32->F16 copy, per-head attention, odd-K matmuls) and fused RoPE and GEGLU. - cmake/laya-ggml.cmake builds ggmlc's laya with TheRock's amdclang against third_party/llama.cpp's ggml and hrx-system; no Vulkan build for the scorer. - serve --laya starts `laya daemon ... --device hrx` with the HSA runtime HRX needs (IREE_HAL_AMDGPU_LIBHSA_PATH from --hrx-libhsa / TheRock). Strix Halo, tests/laya_route_cases.json (200): 95.5% (as the CPU scorer and Vulkan); p50 25 ms a decision, 15-16 ms at a steady 64 tokens (Vulkan 12), 25 ms at 128 (Vulkan 24-31); C++ scorer 538 ms. docs/laya.md also records that the multilingual checkpoint routes worse than typed-decisions in all six languages measured (69.1% vs 83.7%), so it is not used. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * registry: record the llama.cpp pin 22ceda3a (no architecture changes) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: bong-water-water-bong <bong-water-water-bong@1bit.gg> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
This was referenced Oct 7, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
A clean, simple documentation site at https://1bit-monster.github.io/engine/, built from the repository's own
README.mdand everydocs/*.md, so it can't drift from them.tools/site.py(python-markdown, pinned) +site/template.html+site/style.css;.github/workflows/pages.ymlbuilds on PRs and deploys frommain. New files carry the full Apache notice (copyright.pylearns.htmland.css).Reviewed in headless Firefox at 1280 px and 390 px.
🤖 Generated with Claude Code