Multi Repo Qwen Setup
Concrete qwen3-14B guide on NPU against the current worktree's simpler. Points at simpler's own in-repo qwen3_14b_decode example (zero cross-repo), then covers the two cross-repo paths — the pypto-serving runner (prefill/decode TPOT) and the pypto-lib decode_fwd kernel (device decode TPOT + scheduler DFX). Defers cross-repo clone/install to multi-repo-setup, then gives the exact run commands (batch-1 and batch-16), the prefill ring/timeout gotchas that otherwise surface as 507018, how to read the per-token timing layers, and the kernel-level device-timing / swimlane flow. Invoke when running q
Install
npx skills add hw-native-sys/simpler --skill multi-repo-qwen-setupgit clone --depth 1 https://github.com/hw-native-sys/simpler.git /tmp/simpler
mkdir -p ~/.claude/skills
cp -r /tmp/simpler/.claude/skills/multi-repo-qwen-setup ~/.claude/skills/multi-repo-qwen-setupPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Qwen3-14B on NPU: setup + TPOT / overhead measurement This skill is the **qwen3-specific concrete guide**. It first points at simpler's own in-repo qwen3 example (the zero-cross-repo path), then covers the two cross-repo paths against the worktree's simpler: the **pypto-serving runner** (prefill/decode TPOT) and the **pypto-lib `decode_fwd` kernel** (device decode TPOT + scheduler DFX). For the cross-repo paths it gives the exact run commands, the prefill ring/timeout gotchas, and how to read per-token timing — distilled from real runs plus each upstream repo's own docs — sitting on top of [`multi-repo-setup`](../multi-repo-setup/SKILL.md), which owns the generic repo-graph + clone + install steps this skill does not repeat. ## Three ways to run qwen3 — pick by what you're measuring There are **three distinct qwen3 entry points with different invocation styles**. They measure different things; don't confuse them: | aspect | Path 0 (in-repo) | Path A (serving) | Path B (decode_fwd) | | ------ | ---------------- | ---------------- | ------------------- | | repo | simpler (this) | pypto-serving | pypto-lib | | entry | qwen3_14b_decode | npu_generate.py | decode_fwd.py | | cross-repo? | no | yes | yes | | measures | full 40-layer decode correctness/timing | end-to-end TPOT | device decode TPOT + sched DFX | | see | this section | sections 2-5 | section 6 | - **Path 0 — in-repo example (simpler itself, NO cross-repo).** Start here
Details
- Skill name
multi-repo-qwen-setup- Repository
- hw-native-sys/simpler
- Path
.claude/skills/multi-repo-qwen-setup/SKILL.md- Repo updated
- 18 November 2026
FAQ
What does the Multi Repo Qwen Setup skill do?
Concrete qwen3-14B guide on NPU against the current worktree's simpler. Points at simpler's own in-repo qwen3_14b_decode example (zero cross-repo), then covers the two cross-repo paths — the pypto-serving runner (prefill/decode TPOT) and the pypto-lib decode_fwd kernel (device decode TPOT + scheduler DFX). Defers cross-repo clone/install to multi-repo-setup, then gives the exact run commands (batch-1 and batch-16), the prefill ring/timeout gotchas that otherwise surface as 507018, how to read the per-token timing layers, and the kernel-level device-timing / swimlane flow. Invoke when running q
How do I install the Multi Repo Qwen Setup skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add hw-native-sys/simpler --skill multi-repo-qwen-setup. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Multi Repo Qwen Setup popular?
It lives in hw-native-sys/simpler, which has 49 GitHub stars and 102 forks.
Related skills
Benchmark
49hw-native-sys/simpler
Benchmark runtime performance on hardware. If the current branch has commits ahead of upstream/main or uncommitted changes, compares against the fork point (merge-base). Otherwise benchmarks current state only. Use when the user asks to benchmark, measure performance, or compare latency.
Checkout PR
49hw-native-sys/simpler
Add a remote for a given PR's head repository and checkout a local pr-xxx-work branch. Use when the user wants to work on someone else's PR locally.
Clean Branches
49hw-native-sys/simpler
Remove stale local and remote git branches that have been merged into main. Detects regular merges and squash merges. Cleans up fork remote branches and prunes tracking refs. Use when the user asks to clean up branches, remove merged branches, or tidy up git.
Multi Repo Setup
49hw-native-sys/simpler
Set up a cross-repo investigation when a workload from another repo (pypto, pypto-lib, pypto-serving, etc.) needs to be run, especially when you want to swap in simpler-main HEAD or the current worktree's simpler instead of the version that repo pins. Documents the repo graph (who imports who), clones-or-updates each external repo every invocation so stale local clones don't lie about CI parity, and points at each repo's own skills/docs for workload-specific steps. MUST invoke before chasing "X doesn't work on simpler" reports where X lives outside this repo.
Multi Repo Dsv4flash Setup
49hw-native-sys/simpler
Concrete guide to benchmark DeepSeek-V4 (dsv4) FLASH on NPU against the current worktree's simpler — the three decode attentions (swa/csa/hca) and the distributed MoE, all in pypto-lib's models/deepseek/v4/. Defers cross-repo clone/install to multi-repo-setup, then gives the exact PYPTO_BENCH run commands and the hard-won config that makes the numbers reproducible: pin ptoas 0.48 (0.50/ptoas-bin cannot compile csa), pin --start-pos 8192, edit moe.py for 16 experts/card, read fast_effective_us not effective_us, pin an even die, and apply the eager-init idempotency guard so distributed MoE runs
Process Issue
49hw-native-sys/simpler
Process a GitHub Issue as a maintainer, from triage and ownership routing through dependencies and evidence-based closure. Use to manage an Issue's next step; use fix-issue for implementation.
Browse the most popular Claude skills, all git & code review skills, or write your own with the skills generator.