Skip to main content
affpapa.org
RU EN
skill

emerging-techniques-model-pruning — Claude Code skill: what it does

This skill walks through pruning language-model weights with Wanda, SparseGPT, and other methods. Use it to explore model compression without retraining.

Install into Claude Code · claude-code-templates All methods and platforms ↓ 
npx claude-code-templates@latest --skill ai-research/emerging-techniques-model-pruning
Open on GitHub

What emerging-techniques-model-pruning does

The skill covers weight pruning, from magnitude-based removal to Wanda, which uses activation statistics, and SparseGPT, which uses second-order information. It also includes structured pruning and N:M patterns that retain a set number of weights within each group.

Code examples show how to gather activation data, apply pruning masks, and save a model. The examples name torch and transformers as dependencies, and use calibration data with Wanda and SparseGPT. There is a practical trade-off: unstructured sparsity does not provide hardware speedups, while regular N:M patterns are intended for accelerator support.

What it helps with

  • compress language-model weights without retraining
  • rank weights using activation data
  • apply an N:M pattern to model layers
  • save a pruned model

Example prompts for the agent

  • Show me how to prune this language model with Wanda and save the result.
  • Compare magnitude pruning with SparseGPT for my model.
  • Add an N:M structured-pruning example to this code.

How to install emerging-techniques-model-pruning

1claude-code-templates

npx claude-code-templates@latest --skill ai-research/emerging-techniques-model-pruning

2Skills CLI (Vercel)

npx skills add davila7/claude-code-templates --skill emerging-techniques-model-pruning -a claude-code

3Manually: put the files in place

~/.claude/skills/emerging-techniques-model-pruning/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/emerging-techniques-model-pruning/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about emerging-techniques-model-pruning

How do I install emerging-techniques-model-pruning in Claude Code?

Run npx claude-code-templates@latest --skill ai-research/emerging-techniques-model-pruning. Alternatively, add it with npx skills add davila7/claude-code-templates --skill emerging-techniques-model-pruning -a claude-code.

What does emerging-techniques-model-pruning do?

It explains ways to prune language-model weights, including Wanda, SparseGPT, and N:M patterns. The examples cover calibration, mask application, and saving the resulting model.

Which dependencies do the examples use?

The examples use torch and transformers. Wanda and SparseGPT also use calibration data.

Similar skills

All skills for AI and ML →
skill-creator official anthropics Creates and refines skills, then tests them with prompts, evaluations, and performance checks. 414,180installs writing-for-agents Matt Pocock Guidance for writing skills and agent-facing documents. Use it when editing AGENTS.md, CLAUDE.md, or reference files reached from them. 466,719installs media-use heygen-com Finds or generates media for HyperFrames videos, including voiceovers, music, cutouts, and color grades. Stores files locally in assets. CX Codex +1 1,100,000installs mcp-builder official anthropics This skill walks through building an MCP server in TypeScript or Python, from reading the protocol spec to writing tests and evaluation questions. 130,949installs subagent-driven-development Jesse Vincent This skill runs an implementation plan through fresh subagents for independent tasks and reviews each result. Use it for multi-step work in one session. CX Codex 237,746installs orchestration stablyai Coordinates supervised Orca workers with task DAGs, decision gates, and threaded messages. Fits workflows that split work across multiple agents. CX Codex +1 350,300installs

From the same set495

The whole set →
gws-events-renew Renews or reactivates Google Workspace Events subscriptions so they stop expiring under you. Handy when a listener or script depends on a subscription that is about to lapse. 70installs deployment A Claude skill that lets the agent run Railway deployment tasks: read logs, restart, redeploy, take a deploy down. Handy when a service crashes or you need deploy history. 69installs seoagent Runs SEO work inside your repo: audits, keyword clusters, briefs and article drafts land in .seoagent/, so each session picks up where the last one stopped. 69installs gws-cloudidentity A slash command that sends Google Cloud Identity calls through the gws CLI: groups, memberships, devices, SSO profiles. Useful when admin work starts in the chat. 68installs gws-people A Claude Code slash command that drives Google Workspace contacts and profiles through the gws CLI: create, search, update and delete records. 67installs core-components A Claude skill that keeps the agent on your own component library and design tokens instead of raw platform primitives. 66installs cc-skill-coding-standards A skill that hands Claude a fixed set of rules for TypeScript, JavaScript, React and Node.js work: naming, immutability, errors, API shape. 66installs task-execution-engine This skill runs implementation work from Markdown task lists and keeps their status updated. Use it for batch execution, fresh feature work, or a paused run. 65installs

Project sponsors

Companies that keep this analytics open
Gold sponsor
Nutra.Club

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor