# tilekernels-gpu-kernels

This skill guides agents through TileKernels GPU kernels for MoE, quantization and other LLM operations. Use it when writing or checking TileLang code.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 363 installs · ★ 88
- Author: reason-machines
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/reason-machines/trending-skills/tree/HEAD/skills/tilekernels-gpu-kernels/SKILL.md
- Updated: 2026-10-09
- HTML: https://affpapa.org/en/nevibecode/skills/tilekernels-gpu-kernels

## What it does

The skill focuses on TileKernels, a GPU kernel library for LLM workloads. It covers MoE routing, FP8 and FP4 quantization, batched transpose, engram gating and Manifold HyperConnection. It also points to PyTorch-facing wrappers and reference implementations used to validate kernel results.

The covered layout includes `moe`, `quant`, `transpose`, `engram`, `mhc`, `modeling`, `torch` and `testing`. An agent can work with top-k expert selection, token-to-expert mapping, token or block quantization, Sinkhorn normalization and `pytest` checks.

The stated setup requires Python 3.10+, PyTorch 2.10+, TileLang 0.1.9+ and CUDA Toolkit 13.1+. The kernels target NVIDIA GPUs with SM90 or SM100 architecture, so other GPU targets fall outside the stated setup.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill tilekernels-gpu-kernels -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/tilekernels-gpu-kernels/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/tilekernels-gpu-kernels/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill tilekernels-gpu-kernels -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/tilekernels-gpu-kernels/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill tilekernels-gpu-kernels -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/tilekernels-gpu-kernels/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install reason-machines/trending-skills tilekernels-gpu-kernels
```

2. **Manually: put the files in place**
- File: `.github/skills/tilekernels-gpu-kernels/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill tilekernels-gpu-kernels -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/reason-machines/trending-skills --path skills/tilekernels-gpu-kernels
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/tilekernels-gpu-kernels/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add reason-machines/trending-skills --skill tilekernels-gpu-kernels -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/reason-machines/trending-skills/tree/HEAD/skills/tilekernels-gpu-kernels/SKILL.md

## When to use it
- choose kernels for MoE routing
- quantize activations with FP8 or FP4
- transpose batched GPU tensors
- compare a kernel with its PyTorch reference
- run TileKernels tests and benchmarks

## Example prompts
- Use TileKernels to design an MoE route with top-k expert selection.
- Write an FP8 activation quantization path in TileLang and compare it with the PyTorch reference.
- Inspect the `transpose` module and run its `pytest` checks.

## Similar
- [skill-creator](https://affpapa.org/en/nevibecode/skills/skill-creator.md) — skill · 414180 installs
- [writing-for-agents](https://affpapa.org/en/nevibecode/skills/writing-for-agents.md) — skill · 466719 installs
- [media-use](https://affpapa.org/en/nevibecode/skills/media-use.md) — skill · 1100000 installs
- [mcp-builder](https://affpapa.org/en/nevibecode/skills/mcp-builder.md) — skill · 130949 installs
- [subagent-driven-development](https://affpapa.org/en/nevibecode/skills/subagent-driven-development.md) — skill · 237746 installs
- [orchestration](https://affpapa.org/en/nevibecode/skills/orchestration.md) — skill · 350300 installs

From the same set: [reason-machines/trending-skills](https://affpapa.org/en/nevibecode/by/reason-machines/trending-skills.md) — 186 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md