Macaw-OptiQ-4bit

badtheorylabs/Macaw, quantized by OptiQ for Apple Silicon.

On disk2.0 GB

Memoryabout 4 GB

Parent score42.44 LFM2.5-2.6B-OptiQ-4bit

This model reuses the per-layer bit allocation of LFM2.5-2.6B-OptiQ-4bit, whose measured Capability Score is 42.44.

The parent's Capability Score breakdown

This model has no Capability Score of its own. The chart shows the six metrics behind its parent LFM2.5-2.6B-OptiQ-4bit's score of 42.44, taken from the parent's model card.

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/Macaw-OptiQ-4bit

OptiQ Code

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

optiq serve --model mlx-community/Macaw-OptiQ-4bit
optiq code

OptiQ Lab

optiq lab --model mlx-community/Macaw-OptiQ-4bit

Claude Code

optiq run claude -m mlx-community/Macaw-OptiQ-4bit

Codex

optiq run codex -m mlx-community/Macaw-OptiQ-4bit

OpenCode

optiq run opencode -m mlx-community/Macaw-OptiQ-4bit

Mistral Vibe

optiq run vibe -m mlx-community/Macaw-OptiQ-4bit

Hermes Agent

optiq run hermes -m mlx-community/Macaw-OptiQ-4bit

OpenClaw

optiq run openclaw -m mlx-community/Macaw-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/Macaw-OptiQ-4bit")
print(generate(model, tokenizer, prompt="Hello", max_tokens=256))

Details

Family	LFM2.5 · LFM2.5 docs
Architecture	LFM2
Base model	badtheorylabs/Macaw
Size on disk	2.0 GB
Memory needed	about 4 GB
Images	no, text only
Family reference	This model reuses the per-layer bit allocation of LFM2.5-2.6B-OptiQ-4bit, whose measured Capability Score is 42.44.
Downloads	103 in the last month
Updated	2026-09-15
Hugging Face	mlx-community/Macaw-OptiQ-4bit