Inference Perf Bench

Q: What is a Claude Code skill and how does the Inference Perf Bench skill fit in?

A Claude Code skill is a SKILL.md file that lives under ~/.claude/skills/<name>/ and tells the Claude Code CLI agent how to perform a specific task (instructions, prompts, allowed tools). Skills are auto-discovered at session start. Inference Perf Bench is one of 67,000+ skills indexed in the open ClaudSkills catalog, classified under the General category. Learn more at https://claudskills.com/learn/what-is-skill-md/.

Category: General · Sub-category: general-misc · Last updated: 2026-06-15

Canonical inference perf-bench skill (formal name. The colloquial alias is `ai-bench` - identical behaviour). Drives NVIDIA AIPerf + the replay-playback dataset against an in-cluster vLLM endpoint to measure TTFT, ITL, throughput, tok/s/user, request latency, and prefix cache hit rate. Iterative 9-phase workflow. Use when promoting a model to staging/prod, after vllmArgs / vLLM / KV-cache-dtype changes, or for A/B comparison across configs. Triggers on "perf-bench", "AIPerf", "Replay Playback", "throughput sweep", "TTFT P95", "concurrency sweep on inference", "/run-perf-bench", "benchmark throughput", "benchmark latency", "run aiperf", "performance test", or any combination of "perf / latency / throughput / tps / TTFT / ITL" with "inference / vllm / serverless / bench".

About this skill (catalog notes)

Inference Perf Bench includes a dedicated installation section; worked examples; pricing or quota commentary. At roughly 2,205 words the SKILL.md is on the longer end of the catalog distribution.

License: MIT
Original author: cfregly
Indexed lastmod: 2026-06-15
Catalog position: General · general-misc
Indexed related skills: 10

How Inference Perf Bench fits the catalog

Inference Perf Bench sits in the General category under the general-misc sub-topic in the ClaudSkills catalog. There are 10 related skills indexed alongside it; comparing a few before installing usually reveals which fits your workflow best.

These notes are auto-generated from features detected in the SKILL.md file and from this catalog's structure — they aren't part of the source repository.

What this skill does

Inference Perf Bench is a community-contributed Claude Code skill in the general-misc sub-category. It ships as a SKILL.md file that Claude Code auto-discovers under ~/.claude/skills/inference-perf-bench/ and loads when your prompt matches the skill's trigger.

When to invoke it: Use when promoting a model to staging/prod, after vllmArgs / vLLM / KV-cache-dtype changes, or for A/B comparison across configs. Triggers on "perf-bench", "AIPerf", "Replay Playback", "throughput sweep", "TTFT P95", "concurrency sweep on inference", "/run-perf-bench", "benchmark throughput", "benchmark latency", "run aiperf", "performance test", or any combination of "perf / latency / throughput / tps / TTFT / ITL" with "inference / vllm / serverless / bench".

Who uses this skill

The Inference Perf Bench Claude Code skill is built for Claude Code users and developers across all disciplines looking for general-purpose AI assistance. It's part of ClaudSkills (also referred to as Claude Skills or Claude Code Skills) — the open community-curated registry of 117,000+ SKILL.md files for Anthropic's Claude Code agent and the wider Claude ecosystem (Claude API, Claude Agent SDK).

How to install

Free

Manual install (2 steps)

mkdir -p ~/.claude/skills/inference-perf-bench
curl -L https://claudskills.com/skills/inference-perf-bench/SKILL.md \
  -o ~/.claude/skills/inference-perf-bench/SKILL.md

Or just download SKILL.md directly and drop it into ~/.claude/skills/inference-perf-bench/. Claude Code auto-discovers it on next session.

Skills live at ~/.claude/skills/inference-perf-bench/SKILL.md on macOS/Linux, or %USERPROFILE%\.claude\skills\inference-perf-bench\SKILL.md on Windows. See the full install guide for step-by-step instructions.

📱 Install from your phone or desktop Telegram

Open @claudskills_bot on Telegram, tap Open Desktop App, and the desktop app installs this skill for you. Or share the bot link with a colleague — they get the same one-tap install. Learn more →

Pro

One-click install via the desktop app

The ClaudSkills desktop app installs any skill directly into ~/.claude/skills/ with one click — no terminal required. Pro starts at $9/mo or $149 lifetime.

Pro

For the full experience including quality scoring and one-click install features for each skill — upgrade to Pro.

See pricing → Download desktop app

Frequently asked questions

How do I install the Inference Perf Bench Claude Code skill?

Install via the ClaudSkills desktop app (one click) or copy SKILL.md from the source repository to ~/.claude/skills/inference-perf-bench/SKILL.md and restart Claude Code. Both flows are detailed at claudskills.com/install/.

What does the Inference Perf Bench skill do?

Is this skill free to install?

Yes. ClaudSkills is an open registry — every skill keeps its source repository's license, and manual install via copy is free. ClaudSkills Pro ($9/mo, $79/yr, or $149 one-time) adds one-click install via the desktop app and a multi-signal Quality Score.

When should I use the Inference Perf Bench skill?

Use Inference Perf Bench when your Claude Code task falls under the General category — specifically in the general misc area. Claude Code auto-discovers installed skills and invokes the right one based on the task description, so you can also ask Claude directly (e.g. "use Inference Perf Bench" or describe the task and let Claude pick). Browse related skills at /category/general/.

What is a Claude Code skill and how does the Inference Perf Bench skill fit in?

A Claude Code skill is a SKILL.md file that lives under ~/.claude/skills/<name>/ and tells the Claude Code CLI agent how to perform a specific task (instructions, prompts, allowed tools). Skills are auto-discovered at session start. Inference Perf Bench is one of 67,000+ skills indexed in the open ClaudSkills catalog, classified under the General category. Learn more at /learn/what-is-a-claude-skill/.

Attribution & license

Source: https://github.com/cfregly/claude-perf-tune/blob/HEAD/plugins/profile-and-optimize/skills/inference-perf-bench/SKILL.md
License: MIT
Author: cfregly

Cite this skill

If you reference this skill in a blog post, paper, or documentation, you can cite it as:

APA

cfregly. (2026). Inference Perf Bench [Claude Code skill]. ClaudSkills. https://claudskills.com/skills/inference-perf-bench/

BibTeX

@misc{inference-perf-bench-2026,
  author    = {cfregly},
  title     = {Inference Perf Bench [Claude Code skill]},
  year      = {2026},
  publisher = {ClaudSkills},
  url       = {https://claudskills.com/skills/inference-perf-bench/}
}

Embed this skill

Promote, attribute, or link this skill from your own README, blog post, or documentation. All three snippets are free to use — no sign-up, no API key. More distribution surfaces →

Badge

[![ClaudSkills](https://claudskills.com/badge/inference-perf-bench.svg)](https://claudskills.com/skills/inference-perf-bench/?utm_source=badge&utm_medium=readme&utm_campaign=skill_badge)

More General skills

Browse all General skills in the ClaudSkills registry, or explore these other picks from the same category:

Browse all General skills → Top 100 skills

Part of ClaudSkills — the open registry for Claude Skills & Claude Code Skills. · What's New · Install guide · About · llms.txt

Part of Acreator Store — Adam Lankamer's AI tools: PerfectStudio · Ucaption · UTagger · AutoXPoster · TestYourSkills · AutomationFlows · Au Naturel · Telegram @acreatorstore