SynLogic-Mix-3-32B-mlx-OptiQ-5bit

MiniMaxAI/SynLogic-Mix-3-32B, quantized by OptiQ for Apple Silicon.

On disk22.5 GB

Memoryabout 27 GB

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/SynLogic-Mix-3-32B-mlx-OptiQ-5bit

OptiQ Code

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

optiq serve --model mlx-community/SynLogic-Mix-3-32B-mlx-OptiQ-5bit
optiq code

OptiQ Lab

optiq lab --model mlx-community/SynLogic-Mix-3-32B-mlx-OptiQ-5bit

Claude Code

optiq run claude -m mlx-community/SynLogic-Mix-3-32B-mlx-OptiQ-5bit

Codex

optiq run codex -m mlx-community/SynLogic-Mix-3-32B-mlx-OptiQ-5bit

OpenCode

optiq run opencode -m mlx-community/SynLogic-Mix-3-32B-mlx-OptiQ-5bit

Mistral Vibe

optiq run vibe -m mlx-community/SynLogic-Mix-3-32B-mlx-OptiQ-5bit

Hermes Agent

optiq run hermes -m mlx-community/SynLogic-Mix-3-32B-mlx-OptiQ-5bit

OpenClaw

optiq run openclaw -m mlx-community/SynLogic-Mix-3-32B-mlx-OptiQ-5bit

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/SynLogic-Mix-3-32B-mlx-OptiQ-5bit")
print(generate(model, tokenizer, prompt="Hello", max_tokens=256))

Details

Family	Qwen · Qwen docs
Architecture	Qwen2
Base model	MiniMaxAI/SynLogic-Mix-3-32B
Size on disk	22.5 GB
Memory needed	about 27 GB
Images	no, text only
Downloads	26 in the last month
Updated	2026-05-17
Hugging Face	mlx-community/SynLogic-Mix-3-32B-mlx-OptiQ-5bit