Open-weight local LLM

WizardLM 2 (8x22B)

Microsoft AI's ultra-popular fine-tune of Mixtral 8x22B. Apache 2.0 license. Exceptional instruction following and conversational quality.

Large-memory workstation 96 GB RAM Q4_K_M Maximum quality chat
Parameters
8x22B (141B total)
Minimum RAM
96 GB
Model size
88 GB
Quantization
Q4_K_M

Can WizardLM 2 (8x22B) run locally?

WizardLM 2 (8x22B) needs a serious workstation with large unified memory or high VRAM.

Search for wizardlm-2-8x22b in LM Studio or another GGUF-compatible runtime.

chatcodepowerqualitygeneral

Install path

01
Check RAM fitMinimum 96 GB RAM. Start with the Q4_K_M quant.
02
Load the modelSearch wizardlm-2-8x22b in LM Studio.
03
Control locallyUse LocalClaw to manage models, agents, chat, channels and scheduled OpenClaw work.

Strengths

  • Exceptional instruction following
  • Apache 2.0
  • One of the best fine-tunes ever
  • Strong conversational quality

Limitations

  • Requires 96GB+ RAM
  • Very large model
  • Slow inference

Best use cases

  • Maximum quality chat
  • Complex instructions
  • Professional content creation
  • Research

Capability profile

speed
3
quality
10
coding
9
reasoning
9

Technical notes

Developer
Microsoft AI
License
Apache 2.0
Context window
65,536 tokens
Architecture
Mixtral 8x22B fine-tuned with WizardLM pipeline

This model fits these next steps

Hardware fit is based on LocalClaw's RAM tier, model size and quantization metadata. Always leave memory headroom for your OS and runtime.

Similar models to compare

Where to go next