---
id: "eleutherai-lm-evaluation-harness"
title: "lm-evaluation-harness"
creator: "EleutherAI"
creator_url: "https://github.com/EleutherAI/lm-evaluation-harness"
url: "https://github.com/EleutherAI/lm-evaluation-harness"
level: "advanced"
level_max: null
category: "evaluation"
subcategory: "eval-harnesses"
region: "West"
license: "open"
license_scope: "linked-resource"
models: ["any"]
source_date: null
date_added: "2026-10-18"
---

[Evaluation](/evaluation/index.md)

# lm-evaluation-harness

by [EleutherAI](https://github.com/EleutherAI/lm-evaluation-harness)

[Read the original →](https://github.com/EleutherAI/lm-evaluation-harness)

Run model benchmarks through reusable task and backend interfaces.

EleutherAI's harness brings evaluation tasks, model adapters and scoring configuration into a common framework. It helps make comparisons repeatable, while leaving dataset suitability and experimental controls to the evaluator.

Level: [Advanced](/levels/index.md#advanced)
Models: any
Category: [Evaluation](/evaluation/index.md)
Tags: 
Source date: date not stated
Added: 2026-10-18
Last verified: 2026-10-18
Status: ok
Prerequisites: Requires implementation experience or critical reading of technical methods and system trade-offs.
Subcategory: [eval harnesses](/evaluation/eval-harnesses/index.md)
Region: West
Original language: en
Access: free
Resource reuse: open: Source states MIT. Applies to the linked resource, not every referenced model or dataset. [Facet definitions](/about/index.md#facets)

Summary in our own words. Rights remain with the original creator. [Credit and correction policy](/about/index.md#credit-policy).

[Plain-text Markdown copy](/evaluation/eval-harnesses/eleutherai-lm-evaluation-harness/index.md)