Individual skill
batch-triton-ascend-token-unreorder
Batch inverse MoE token reordering in Triton-Ascend with contiguous tile loads, extension.extract_slice row extraction, and indexed output stores. Use when an agent must scatter sequential routed tokens back to output[indices[j]], reduce excessive one-token logical cores, or exploit contiguous reads while accepting dispersed writes on Ascend NPU.
skills/batch-triton-ascend-token-unreorder/SKILL.md
Source commit 4ab5ee7adf47755ecd42723f55f4c58e38be95a7
What was checked
- Matches the observed repository HEAD
- Yes
- Scanner rules are current
- No
- Repository scan complete
- No
- Skill scan complete
- No
- Observed behavior
- Not checked yet
- Last checked
- Nov 4, 2026
These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.
Install this skill
Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.
An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.
Repository context
- Repository stars (not skill ratings)
- 189
- Repository last updated
- Aug 15, 2026
- License
- Apache-2.0
Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.
- ai Repository topic
- ascend Repository topic
- ascend-npu Repository topic
- npu Repository topic
- triton Repository topic
- triton-ascend Repository topic
- claude-code Repository topic
- codex Repository topic
- opencode Repository topic
- engineering Derived category
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Read excerpt
# Batch Triton-Ascend Token Unreorder
## Goal
Read a contiguous block of routed token rows once, extract rows in UB, and scatter each row to its indexed destination.
## Workflow
1. Treat input as S by D contiguous routed rows and indices as destination rows.
2. Assign BLOCK_SIZE consecutive input rows to each program.
3. Load a BLOCK_SIZE by D input tile and BLOCK_SIZE indices with masks.
4. Extract each valid row with extension.extract_slice.
5. Store the row at indices[row] * D with a column mask.
