# AI Leaderboard

> A public table that ranks AI models by test scores or votes. The order changes often.

**Field:** Artificial Intelligence  
**Category:** Evaluation & Benchmarks  
**Also known as:** LLM Leaderboard, Model Leaderboard

## Think of it like

It is like a sports league table. Teams move up and down as games are played.

## Definition

An AI leaderboard is a public table that ranks models by their scores. Some boards use one benchmark. Others mix many tests, or use votes from people. Boards are updated as new models come out, so the order changes often. Each board measures its own thing, such as coding, math or chat. A model that is first on one board can be far lower on another.

## Example

Before picking a coding model, Emma checks two boards. One ranks by SWE-bench scores and one by people's votes. She tries the models that do well on both.

## Why it matters

It is the quickest way to see which models to try. But it is a place to start, not a final answer.

## Common mistake

Picking the top model without reading what the board measures. Your task may not be what the board tests.

## Three principles

- **Ranked table.** Scores or votes, in order.
- **Always moving.** New models change the order.
- **One board, one skill.** Read what it measures.

## Related terms

- [AI Benchmark](https://learn.benibix.com/term/ai/ai-benchmark)
- [LMArena (Chatbot Arena)](https://learn.benibix.com/term/ai/chatbot-arena)
- [SWE-bench](https://learn.benibix.com/term/ai/swe-bench)
- [Benchmark Contamination](https://learn.benibix.com/term/ai/benchmark-contamination)
- [Benchmark Saturation](https://learn.benibix.com/term/ai/benchmark-saturation)
- [Model Selection (LLM)](https://learn.benibix.com/term/ai/model-selection)

## Told apart from

- [AI Benchmark vs AI Leaderboard](https://learn.benibix.com/compare/ai-benchmark-vs-ai-leaderboard): People use benchmark and leaderboard as if they meant one thing. A benchmark is the test: fixed questions and a fixed way to score them. A leaderboard is the public table that ranks models by their scores. One leaderboard can mix many benchmarks. LMArena ranks models by people's votes, with no fixed test at all.

![Two ranked tables side by side, one for coding and one for chat. The same four models appear on both, and model C is first on one board and third on the other.](https://learn.benibix.com/diagrams/ai5-leaderboard.svg)

---

Source: https://learn.benibix.com/term/ai/ai-leaderboard
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-19
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
