# zjunlp/EasyEdit

[ACL 2024] An Easy-to-use Knowledge Editing Framework for LLMs.

Repository: https://github.com/zjunlp/EasyEdit
Canonical: https://ross.abutalabs.com/products/easyedit
Homepage: https://zjunlp.github.io/project/KnowEdit
Language: Jupyter Notebook
License: MIT
License Family: permissive
Topics: artificial-intelligence, efficient, knowledge-editing, large-language-models, model-editing, natural-language-processing, tool, easyedit, knowlm, gpt, llama, baichuan, chatgpt, mmedit, unlearning, trustworthy-ai, safeedit, cknowedit, knowedit, easyedit2
Last push: 2026-07-14T12:47:07+00:00

## Health v2 (maintenance only)
Score: 71/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 92, release rhythm 35, longevity 86
- inputs: {"age_days": 1212, "days_push": 50, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2908, forks 372 (observed 2026-08-28T04:07:29.113958+00:00)

## What it is
EasyEdit is an easy-to-use knowledge editing framework for large language models, supporting methods that modify model parameters or steer behavior at inference time. It includes benchmarks like KnowEdit and supports models such as GPT, LLaMA, and Baichuan.

## Use cases
- edit factual knowledge in an LLM without retraining
- unlearn harmful or outdated knowledge from a model
- steer LLM behavior at inference time
- benchmark knowledge editing methods on KnowEdit
- update wrong facts in a fine-tuned llama model
- research model editing techniques for trustworthy AI

## When to choose
- you need to correct or update specific facts in an open-source LLM
- you are researching knowledge editing or model editing methods
- you need machine unlearning or safety steering for LLMs

## When to avoid
- you need full model retraining or fine-tuning pipelines
- you only need prompt engineering without model modification
- you work with closed-source APIs that don't expose model weights

## Facets
- artifact type: framework
- maturity: active
- function: machine-learning, llm-inference, llm-training, prompt-engineering, developer-tools
- domain: large-language-models, artificial-intelligence, machine-learning
- platform: python, cross-platform
- tags: knowledge-editing, model-editing, llm-steering, unlearning, trustworthy-ai, knowedit, safeedit, research-tool, natural-language-processing, gpu

## Member repositories
- zjunlp/EasyEdit (main) score 71

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:29.113958+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:34:22.496276+00:00, confidence not recorded.
  - readme: https://github.com/zjunlp/EasyEdit (fetched 2026-08-28T04:07:29.113958+00:00, sha 7606ca124117)
  - homepage: https://zjunlp.github.io/project/KnowEdit (fetched 2026-08-29T09:49:35.793559+00:00, sha 983eaef7b39f)
- Data as of 2026-08-30T08:39:29.467469+00:00.