LLM locally on iOS and MacOS
Absolutely free, open source and private.
Download Latest From TestFlightDownload Stable From AppStoreAbsolutely free
LLM Farm provides all features absolutely free of charge! No hidden fees, subscriptions or feature limitations - all features are available for use at no additional cost.
Open Source
The core is a Swift library based on llama.cpp, ggml and other open source projects that allows you to perform various inferences. A class hierarchy has been developed that allows you to add your own inference.Features
Supported models
Support many populsar models and many others
LLaMa
LLaMa 1, 2, 3, 3.1, 3.2 Popular models from Meta.
Phi
Popular models from Microsoft is a 3.8 billion-parameter, lightweight, state-of-the-art open model trained using the Phi-3 datasets
Gemma
Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models.
MobileVLM
MobileVLM is a competent multimodal vision language model (MMVLM) targeted to run on mobile devices.
Qwen
Popular Chinese models.
Bunny
Bunny is a family of lightweight but powerful multimodal models. It offers multiple plug-and-play vision encoders, like EVA-CLIP, SigLIP and language backbones, including Llama-3-8B, Phi-3-mini, Phi-1.5, StableLM-2, Qwen1.5, MiniCPM and Phi-2.
TinyLlama
The TinyLlama project aims to pretrain a 1.1B Llama model on 3 trillion tokens.
Mamba
Mamba is a new state space model architecture showing promising performance on information-dense data such as language modeling, where previous subquadratic models fall short of Transformers.
Santacoder
The SantaCoder models are a series of 1.1B parameter models trained on the Python, Java, and JavaScript subset of The Stack (v1.1) (which excluded opt-out requests).
RWKV
RWKV (pronounced RwaKuv) is an RNN with GPT-level LLM performance, and can also be directly trained like a GPT transformer (parallelizable). We are at RWKV v6.
Other models
See full list here.
