skills.rest
raghavbadhwar
raghavbadhwarCOMMUNITY
Explore all skills by this author →
GitHub

benchmark-models

Benchmarks Claude, GPT, and Gemini on the same prompt comparing latency, cost, and quality.

Updated Aug 9, 2026
Search Tags:#claude#latency#gemini#gpt#llm comparison#llm judge#model benchmark
One-click install
npx skills add https://github.com/raghavbadhwar/rstack --skill benchmark-models-raghavbadhwar
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: benchmark-models
Source: https://github.com/raghavbadhwar/rstack/tree/main/benchmark-models
Command: npx skills add https://github.com/raghavbadhwar/rstack --skill benchmark-models-raghavbadhwar

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Choosing between AI models for a given task is usually guesswork. This Skill runs the same prompt through Claude, GPT (via Codex CLI), and Gemini side-by-side and reports latency, token usage, cost, and optional quality scores so model selection is based on measured data. ## Core Features & Use Cases - Cross-model execution: Sends one prompt to multiple providers and compares speed, tokens, and cost in a single table. - LLM quality judge: Optionally scores each model's output on a 0-10 scale using Claude as a tiebreaker judge. - Skill benchmarking: Can benchmark an existing gstack skill's SKILL.md as the prompt to expose tool-use differences between models. - Use Case: You are deciding which model should run your code-review workflow. Run the benchmark against your review skill and compare each model's latency, cost, and judged output quality before committing. ## Quick Start Ask the agent to run a cross-model benchmark comparing Claude, GPT, and Gemini on a chosen prompt or skill file.

Frequently Asked Questions about benchmark-models

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I compare Claude, GPT, and Gemini on the same prompt?▼

Run the benchmark-models workflow, which sends one prompt to all three providers and prints a side-by-side table of latency, token usage, and cost. You can supply the prompt inline, as a file, or by picking an existing skill.

How do I benchmark an AI model on my own skill or prompt file?▼

Choose the option to benchmark an existing skill and the workflow lists available SKILL.md files to use as the prompt. Alternatively, pass a path to any prompt file on disk and it will be sent verbatim to each provider.

What happens if one model provider is not authenticated?▼

A dry-run check shows which providers are ready and which are not, with remediation hints. Unauthenticated providers are skipped cleanly without aborting the batch, but at least one authed provider is required to run.

Does the benchmark measure output quality or just speed and cost?▼

By default it measures latency, tokens, and cost. If an Anthropic API key is available, you can enable an LLM judge that scores each model's output on a 0-10 scale for a small additional cost per run.

What is the difference between benchmark-models and the benchmark skill?▼

benchmark-models measures AI model performance on prompts or skills across providers. The separate benchmark skill measures web page performance such as Core Web Vitals and load times.

Related Skills

Explore More Software Engineering Skills→
cheng80cheng80
1

benchmark-models

Benchmark AI models across Claude, GPT, and Gemini for latency, token usage, and cost.

Community
Advanced
PSkinnerTechPSkinnerTech

ai-evals

Design and run AI evaluations to measure model quality.

Community
Advanced
surus-latsurus-lat
8

run-benchmark

Execute end-to-end AI benchmarking workflows with spec validation and smoke testing.

Official
Advanced
thejaminatorthejaminator
32

compare-models

Compare multiple LLM models side-by-side via CLI with structured output.

Community
Intermediate