mlx-optiq
← Models

Qwen3.6-27B-OptiQ-4bit

Qwen/Qwen3.6-27B, quantized by OptiQ for Apple Silicon; takes images.

visionMTP head
On disk20.0 GB
Memoryabout 24 GB
Capability Score86.04
Imagesyes

Capability Score breakdown

The six metrics behind this model's Capability Score of 86.04, as reported in the model card. Values are compared with uniform 4-bit.

  • This model
  • Uniform 4-bit
Capability Score breakdown: Qwen3.6-27B-OptiQ-4bitMMLU: This model 87.4, Uniform 4-bit 87.6; GSM8K: This model 92.0, Uniform 4-bit 92.1; IFEval: This model 74.1, Uniform 4-bit 71.7; BFCL-V3: This model 92.5, Uniform 4-bit 93.0; HumanEval: This model 90.2, Uniform 4-bit 92.1; HashHop: This model 80.0, Uniform 4-bit 77.0MMLU87.487.6GSM8K92.092.1IFEval74.171.7BFCL-V392.593.0HumanEval90.292.1HashHop80.077.0

Use it

Install

pip install mlx-optiq

Serve it

An OpenAI- and Anthropic-compatible server on this Mac. Every app below can use it.

optiq serve --model mlx-community/Qwen3.6-27B-OptiQ-4bit

OptiQ Code

The terminal coding agent uses whatever model optiq serve is running.

optiq serve --model mlx-community/Qwen3.6-27B-OptiQ-4bit
optiq code

OptiQ Lab

optiq lab --model mlx-community/Qwen3.6-27B-OptiQ-4bit

Claude Code

optiq run claude -m mlx-community/Qwen3.6-27B-OptiQ-4bit

Codex

optiq run codex -m mlx-community/Qwen3.6-27B-OptiQ-4bit

OpenCode

optiq run opencode -m mlx-community/Qwen3.6-27B-OptiQ-4bit

Mistral Vibe

optiq run vibe -m mlx-community/Qwen3.6-27B-OptiQ-4bit

Hermes Agent

optiq run hermes -m mlx-community/Qwen3.6-27B-OptiQ-4bit

OpenClaw

optiq run openclaw -m mlx-community/Qwen3.6-27B-OptiQ-4bit

optiq run starts optiq serve if it is not running, sets the agent up for this session only, and stops the server when the agent exits.

mlx-lm, text only

from mlx_lm import load, generate

model, tokenizer = load("mlx-community/Qwen3.6-27B-OptiQ-4bit")
print(generate(model, tokenizer, prompt="Hello", max_tokens=256))

Details

FamilyQwen · Qwen docs
ArchitectureQwen3.5
Base modelQwen/Qwen3.6-27B
Size on disk20.0 GB
Memory neededabout 24 GB
Imagesyes, through optiq serve and the Lab
MTP headincluded
Capability Score86.04 (from the model card)
Downloads3,304 in the last month
Updated2026-10-18
Hugging Facemlx-community/Qwen3.6-27B-OptiQ-4bit