mlx-optiq
← Models

MN-Crimson-Cafe-12B-OptiQ-4bit

EldritchLabs/MN-Crimson-Cafe-12B, quantized with OptiQ by jaithree for Apple Silicon.

community
On disk9.1 GB
Memoryabout 12 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 jaithree/MN-Crimson-Cafe-12B-OptiQ-4bit

OptiQ Code

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

optiq serve --model jaithree/MN-Crimson-Cafe-12B-OptiQ-4bit
optiq code

OptiQ Lab

optiq lab --model jaithree/MN-Crimson-Cafe-12B-OptiQ-4bit

Claude Code

optiq run claude -m jaithree/MN-Crimson-Cafe-12B-OptiQ-4bit

Codex

optiq run codex -m jaithree/MN-Crimson-Cafe-12B-OptiQ-4bit

OpenCode

optiq run opencode -m jaithree/MN-Crimson-Cafe-12B-OptiQ-4bit

Mistral Vibe

optiq run vibe -m jaithree/MN-Crimson-Cafe-12B-OptiQ-4bit

Hermes Agent

optiq run hermes -m jaithree/MN-Crimson-Cafe-12B-OptiQ-4bit

OpenClaw

optiq run openclaw -m jaithree/MN-Crimson-Cafe-12B-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("jaithree/MN-Crimson-Cafe-12B-OptiQ-4bit")
print(generate(model, tokenizer, prompt="Hello", max_tokens=256))

Details

FamilyMistral · Devstral · Mistral · Devstral docs
ArchitectureMistral
Base modelEldritchLabs/MN-Crimson-Cafe-12B
Size on disk9.1 GB
Memory neededabout 12 GB
Imagesno, text only
Downloads17 in the last month
Updated2026-08-10
Hugging Facejaithree/MN-Crimson-Cafe-12B-OptiQ-4bit