Mistral · 46.7B · Desktop / import

Mixtral 8x7B Instruct on a phone

Mixture-of-experts — high quality without always loading a dense 70B.

Parameters
46.7B
Typical Q4
~26 GB (Q4) / less active
RAM to budget
~32 GB
Engines
GGUF

Get it running

Sparse MoE; ~12–26 GB depending on quant; LM Studio classic. Hugging Face: mistralai/Mixtral-8x7B-Instruct-v0.1.

Load in LM Studio or Ollama, then chat from the phone.

Get LM Mini and load a model that actually fits — on-device, or via Connect to a PC.

Phones that can run it

  • None in this guide — use a Mac or Connect.

Tight (close other apps)

  • None in this guide — use a Mac or Connect.