MergeKit
A command-line toolkit for merging the weights of pretrained large language models.
These buttons open the developer's own site, repository or store listing in a new tab. wares.gg does not host downloads.
About MergeKit
MergeKit combines two or more pretrained language models into a single model. It supports many merging algorithms, so you can experiment with different ways of blending weights without training from scratch.
It works out of core, which keeps memory use low when merges get elaborate. Merges can run entirely on a CPU, or on a GPU with as little as 8 GB of VRAM. It suits people who build and publish fine-tuned or merged open-weight models.
Strengths
- Supports many model merging algorithms
- Out-of-core approach keeps memory use low
- Runs on CPU or on a GPU with 8 GB of VRAM
Limitations
- Command-line and config driven, with no graphical interface
- Useful mainly to people already working with model weights
Details
- Pricing
- FreeFree to download and run from its GitHub repository.
- License
- Proprietary
- Developer
- Arcee AI
- Platforms
- Command line
- How it runs
- Downloadable app
- Account
- Not required
- Works offline
- Yes
- Best suited for
- Building merged open-weight language models on modest hardware
- Categories
- Local AI tools, AI developer tools
- Last verified
- Added
- Sources
Similar software
Related functionality, not necessarily a direct replacement.
LlamaFactory
Web UI and command-line toolkit for fine-tuning over 100 open language and vision-language models.
Axolotl
A command-line tool for fine-tuning language models from YAML configuration files.
Unsloth
A local interface and command-line tool for running and fine-tuning language and diffusion models.
Hugging Face Hub
A web platform for hosting, finding and sharing machine learning models, datasets and demo apps.
lm-evaluation-harness
EleutherAI's command-line framework for few-shot evaluation of language models on many benchmark tasks.
