Mlx Bench

Name: Mlx Bench
Author: cavi-ai

ASecurity

Measure a locally served MLX model without downloading it.

2 stars
0 votes
0 copies
1 view
Added October 3, 2026

ai-agentspythonshell

Security analysis

A100/100

Scanned October 3, 2026

npx -y skills add cavi-ai/mlx-agent --skill mlx-bench --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Mlx Bench?

Add the live security badge to your README. It updates with every re-scan.

[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/cavi-ai-mlx-bench-aa868404/badge)](https://www.skillsdirectory.com/skills/cavi-ai-mlx-bench-aa868404)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro

SKILL.md

---
name: "mlx-bench"
description: "Measure a locally served MLX model without downloading it."
---

# MLX Bench

Resolve `<plugin-root>` as the absolute directory two levels above the directory containing this SKILL.md; it holds `scripts/` and `src/`. Never resolve the bundled executable from the shell working directory.

canonical capability ID: mlx-agent.bench

Measure a model already served by a running local runtime:

`python3 <plugin-root>/scripts/mlx-agent bench run --repo <repo> --runtime <runtime> --json`

Present the returned measurements as returned. Bench must not start servers, download model weights, or change configuration; the model must already be served by an existing local runtime.