LLM Farm 4+

Name: LLM Farm
Rating: 3.3 (6 reviews)
Author: Artem Savkin

Artem Savkin

Entwickelt für iPad

- 3,3 • 6 Bewertungen

- Gratis

Beschreibung

LLMFarm is an iOS and MacOS app to work with large language models (LLM). It allows you to load different LLMs with certain parameters.

# Features
* Various inferences
* Various sampling methods
* Metal
* Model setting templates

# Inferences
* LLaMA
* GPTNeoX
* Replit
* GPT2 + Cerebras
* Starcoder(Santacoder)
* RWKV
* Falcon
* MPT
* Bloom
* StableLM-3b-4e1t
* Qwen
* Gemma
* Phi
* Mamba
* Others

# Multimodal
* LLaVA 1.5 models
* Obsidian
* Bunny
* MobileVLM 1.7B/3B models

Note: For Falcon, Alpaca, GPT4All, Chinese LLaMA / Alpaca and Chinese LLaMA-2 / Alpaca-2, Vigogne (French), Vicuna, Koala, OpenBuddy (Multilingual), Pygmalion/Metharme, WizardLM, Baichuan 1 & 2 + derivations, Aquila 1 & 2, Mistral AI v0.1, Refact, Persimmon 8B, MPT, Bloom select llama inferece in model settings.

Sampling methods
* Temperature (temp, tok-k, top-p)
* Tail Free Sampling (TFS)
* Locally Typical Sampling
* Mirostat
* Greedy
* Grammar

15. Okt. 2024

Version 1.3.9

Changes:
* llama.cpp updated to b3837
* Added support for llama 3.2, 3.1, RWKV, MiniCPM(2.5, 2.6, 3), Chameleon models
* Added Metal support for Mamba
* Added Llama 3.2 download links (working with LLaMa3 Instruct template)
* Added Phi 3.5 download links (working with Phi 3 template)
* Added Bunny template and download link
* Some metal improvements
* Fixed some errors for gemma2, DeepSeek-V2, ChatGLM4, llama 3.1, T5, TriLMs, BitNet models
* Fixed some other errors

3,3 von 5

6 Bewertungen

Privacy…

Do the chats leave my ipad and do you collect chats to train some other AI or for some other things?

No.

Simple but fast and works

It is not remotely as refined as LM studio (it is pretty barebone) but on an ipad with M1 chip (and if you activate Metal) it runs almost all smaller LLMs fast and reliable. It is not suitable for developing projects in it since you cannot really edit your input history, only delete it, but it is great as an offline mobile environment for short LLM sessions.The quality of output of course depends on the LLMs you install. I recommend smaller LLMs since the ipads capacities, though astonishing, are limited.
The best thing is you can make it run almost all smaller LLMs LM studio can run and download.
Thank you for publishing this.

Thanks for the feedback. I will try to add story editing features in the next versions.

Won’t download model

The app can’t even complete the download of a llama model. Seems like that’s a large hurdle to using the app.

Der Entwickler, Artem Savkin, hat darauf hingewiesen, dass die Datenschutzrichtlinien der App den unten stehenden Umgang mit Daten einschließen können. Weitere Informationen findest du in den Datenschutzrichtlinien des Entwicklers.