mamay-12b-mlx-optiq_4bit

shportix/MamayLM-Gemma-3-12B-IT-W4A16, quantized with OptiQ by mstyslavity for Apple Silicon.

community

On disk7.1 GB

Memoryabout 10 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 mstyslavity/mamay-12b-mlx-optiq_4bit

OptiQ Code

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

optiq serve --model mstyslavity/mamay-12b-mlx-optiq_4bit
optiq code

OptiQ Lab

optiq lab --model mstyslavity/mamay-12b-mlx-optiq_4bit

Claude Code

optiq run claude -m mstyslavity/mamay-12b-mlx-optiq_4bit

Codex

optiq run codex -m mstyslavity/mamay-12b-mlx-optiq_4bit

OpenCode

optiq run opencode -m mstyslavity/mamay-12b-mlx-optiq_4bit

Mistral Vibe

optiq run vibe -m mstyslavity/mamay-12b-mlx-optiq_4bit

Hermes Agent

optiq run hermes -m mstyslavity/mamay-12b-mlx-optiq_4bit

OpenClaw

optiq run openclaw -m mstyslavity/mamay-12b-mlx-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("mstyslavity/mamay-12b-mlx-optiq_4bit")
print(generate(model, tokenizer, prompt="Hello", max_tokens=256))

Details

Family	Gemma-4 · Gemma-4 docs
Architecture	Gemma 3
Base model	shportix/MamayLM-Gemma-3-12B-IT-W4A16
Size on disk	7.1 GB
Memory needed	about 10 GB
Images	no, text only
Downloads	30 in the last month
Updated	2026-05-17
Hugging Face	mstyslavity/mamay-12b-mlx-optiq_4bit