Mixtral 8x7B (MoE)

Name: Mixtral 8x7B (MoE)
Author: Mistral (Ollama)

Scores reviewed 2026-06-16

Mistral (Ollama)

LMFinder id: ollama/mixtral-8x7b

Context: 32K
Price per 1M: Free
Tooling: 3/5
Deploy: Local (self-hosted)

Best for

Writing and editing code
Complex reasoning and analysis
Private, offline use on your own hardware

Skip if

You don't have a 48GB+ GPU
You want a managed, zero-setup API

Capability scores

Tool calling3/5

Tooling (1-5): how reliably the model calls tools and follows multi-step instructions. Editorial score from LMFinder, not a benchmark.

Multi-step reasoning3/5

Specifications

Context window: 32K tokens
VRAM: ~48GB
License: apache-2.0
Pricing: Free to run · needs ~48GB GPU
Deployment: Local (self-hosted)
Docs: Model docs ↗

Our take

Mixtral 8x7B (MoE) from Mistral (Ollama) offers a 32K-token context window and is well-suited to coding. It runs on your own hardware, so there is no per-token cost and your data stays private (plan for roughly 48GB of VRAM).

Watch for the trade-offs: there is no managed API — you run and maintain it yourself.

Compare to

Gemma 3 27B

Google (Ollama)

128K context · Free to run · needs ~16GB GPU

Llama 3.1 70B

Meta (Ollama)

128K context · Free to run · needs ~40GB GPU

Mistral Nemo 12B

Mistral (Ollama)

128K context · Free to run · needs ~8GB GPU

Best Local LLMs for Hermes Agent →Local & Open-Weight LLMs →Best Local LLMs for Ollama →

Not sure this is the one?

Describe your project and let LMFinder rank the field for your exact constraints.

Describe your project →