bench

Official

Benchmark API model latency and throughput.

AuthorOpenDCAI
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Quick benchmark of third-party API models for latency and throughput to help developers compare production-ready performance across endpoints.

Core Features & Use Cases

  • Two Python scripts (test_models.py and test_anthropic.py) perform streaming-based benchmarks against different model endpoints.
  • Automatically discover available models from /models and filter to the latest mainstream text models (Claude 4.x, GPT-5.x, Gemini-3.x, Qwen 3.x, GLM 4.7+, Kimi k2.5+), excluding outdated or irrelevant variants.
  • Collect and compare metrics such as TTFB, total time, generated token count, and tokens-per-second to enable fast model selection.
  • Supports multiple interface formats (OpenAI-like /v1/chat/completions and Anthropic /v1/messages) for cross-compatibility.

Quick Start

Use a concise instruction to start the benchmark by providing the API key and base URL for the endpoint.

Dependency Matrix

Required Modules

aiohttp

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: bench
Download link: https://github.com/OpenDCAI/leonai/archive/main.zip#bench

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.