llama.cpp

WindowsMacLinuxFreeVisit site ↗

The engine many local apps use to run GGUF files. The project documents the commands.

Models that run in llama.cpp

Gemma 4 31B
great at writing text
WritingChatArchiveTranslate
Runs in:
20 GB · Free · Apache 2.0
Qwen3.8 27B
excellent at writing, code, and translation
WritingChatArchiveCodeTranslate
Runs in:
Fits on:
- Mac 32GB+
18 GB · Free · Apache 2.0
Gemma 4 26B A4B
great at writing text, and quicker than it looks
WritingChatArchiveTranslate
Runs in:
Fits on:
- Mac 32GB+
16 GB · Free · Apache 2.0
Qwen3 235B-A22B
a giant chatbot for the biggest computers
WritingChatCode
Runs in:
120 GB · Free · Apache 2.0
Qwen3.5 27B
strong at writing, code, and other languages
WritingChatArchiveCodeTranslate
Runs in:
Fits on:
- Mac 32GB+
18 GB · Free · Apache 2.0
Qwen3.5 35B-A3B
fast chat and code, because only a slice wakes up
WritingChatCode
Runs in:
20.4 GB · Free · Apache 2.0
Gemma 3 27B
a strong writer, a step behind Gemma 4
WritingChatArchive
Runs in:
Fits on:
- Mac 32GB+
17 GB · Free · Gemma Terms of Use
Qwen3-Coder 30B-A3B
fast help with code
Code
Runs in:
Fits on:
- Mac 32GB+
19 GB · Free · Apache 2.0
gpt-oss 120B
OpenAI's big open model, for huge memory pools
WritingChatCode
Runs in:
65 GB · Free · Apache 2.0
Gemma 3 12B
great at writing text on a 16 GB machine
WritingChatArchiveTranslate
Runs in:
Fits on:
- Mini PC
- Laptop 16GB
8.2 GB · Free · Gemma Terms of Use
DeepSeek R1 Distill 32B
talks through the steps before it answers
WritingChatCode
Runs in:
20 GB · Free · MIT
Qwen3 8B
a quick chatbot that fits smaller computers
WritingChatArchiveCodeTranslate
Runs in:
Fits on:
- Mini PC
- Laptop 16GB
5.2 GB · Free · Apache 2.0
Gemma 3n E4B
pleasant everyday notes on a laptop
WritingChat
Runs in:
Fits on:
- Mini PC
- Laptop 16GB
5 GB · Free · Gemma Terms of Use
Llama 3.3 70B
a famous big chatbot that needs lots of memory
WritingChatArchive
Runs in:
40 GB · Free · Llama 3.3 Community License
gpt-oss 20B
a solid helper for chat and code
WritingChatCode
Runs in:
Fits on:
- Mac 32GB+
- Workstation 24GB GPU
14 GB · Free · Apache 2.0
Gemma 3 4B
great at writing text, and it can look at a photo
WritingChatTranslate
Runs in:
Fits on:
- Raspberry Pi 5
- Phone
3.4 GB · Free · Gemma Terms of Use
Qwen2.5-Coder 32B
a well-known local coding helper
Code
Runs in:
20 GB · Free · Apache 2.0
Aya Expanse 32B
a multilingual writer
WritingChatTranslate
Runs in:
20 GB · Free · CC-BY-NC 4.0
Phi-4 14B
careful answers, strong for its size
WritingChatCode
Runs in:
Fits on:
- Mini PC
- Laptop 16GB
9.1 GB · Free · MIT
Qwen2.5-Coder 7B
helps with short bits of code
Code
Runs in:
Fits on:
- Raspberry Pi 5
- Mini PC
4.7 GB · Free · Apache 2.0
Aya Expanse 8B
writes and translates across languages
WritingChatTranslate
Runs in:
Fits on:
- Mini PC
- Laptop 16GB
5 GB · Free · CC-BY-NC 4.0
Llama 3.2 3B
great at short rewrites on a tiny computer
WritingChat
Runs in:
Fits on:
- Raspberry Pi 5
- Phone
2 GB · Free · Llama 3.2 Community License
Qwen2.5-VL 7B
reads text from scans (OCR) and answers about the picture
Scan
Runs in:
Fits on:
- Mini PC
- Laptop 16GB
6 GB · Free · Apache 2.0

Models that run in llama.cpp

Other apps that run these models