mergekit

End to End Generative AI Industry Projects on LLM Models with Deployment_Awesome LLM Projects

All-in-one UI for merged LLMs in Hugging Face

Tools for merging pretrained Large Language Models and create Mixture of Experts (MoE) from open-source models.

A fast Rust CLI for LLM model merging, diffing, and conversion. 10 merge algorithms, mergekit-compatible configs, safetensors native.

Conflict-free merge for DataFrames, JSON, ML models & distributed agents — powered by CRDTs. The first merge library where every operation is mathematically guaranteed to converge.

Mergekit supports various architectures like Llama, Mistral, and more models, enabling merges on CPU or GPU with low memory needs through lazy tensor loading and out-of-core processing. It handles elaborate operations such as interpolated gradients, frankenmerging (layer-wise), LoRA extraction, and mixture of experts merging.​ [ by Arcee AI ]

Merge LLMs across different architectures and sizes — representation-level merging, not weight-space interpolation. Merge GPT-2, OPT, DistilGPT-2, SmolLM2, and more

Intelligent layer pruning toolkit for LLMs featuring iterative optimization, self-healing algorithms, and comprehensive benchmarking.

A continuous generalist web navigation agent that merges individually fine-tuned LLMs as domain experts.

Mergekit Assistant is a cutting-edge toolkit designed for the seamless merging of pre-trained language models. It supports an array of models, offers various merging methods, and optimizes for low-resource environments with both CPU and GPU compatibility.

A visual node-based pipeline studio for local AI models — uncensor, merge, compress, and deploy without writing code.

Reward engineering, GRPO fine-tuning and model merging for Turkish multiple-choice question answering with GPT-2-based language models.

First SLERP merge of Mamba-2 hybrid LLMs (Falcon-H1-7B-Instruct × H1R-7B). Includes merge script, benchmarks, and architecture documentation.

Version control for what a model has learned. Ballast stores weight deltas as content-addressed blocks, records merges as views over their inputs, and deletes with a proof.

Spark 2.5 4B burger merge: passthrough sammich + evolutionary search, merge-only

An artificial-life arena where open-weight LLMs survive, reproduce by weight-merge, and evolve. Open-science AI-safety sandbox + survival benchmark for open models + a watchable show. Idea & design by Andrii Shramko.

Merge language models directly in your browser without code. Run merges on your CPU or GPU and download your combined models.