OptiQ quants, ready to load.
Every model is a standard MLX checkpoint. Load it with mlx_lm.load(...), no special runtime and no PyTorch. OptiQ measures each layer's sensitivity and spends bits where they matter, so a mixed-precision quant keeps the accuracy a flat uniform quant gives up, at the same size.
import optiq # unlocks every OptiQ arch from mlx_lm import load, generate model, tok = load( "mlx-community/Laguna-XS-2.1-OptiQ-4bit") print(generate(model, tok, prompt="Hi"))
$ optiq convert Qwen/Qwen3.6-27B --target-bpw 4.5Pick a family. Details in the docs.
Each family page lists every quant, its size, and its six-metric Capability Score against stock uniform 4-bit. Here is the lineup, newest first.
Muse-Glimmer
A 30B image-text reasoning model, Capability Score 87.36. Sliding-window attention on three layers in four keeps memory flat as context grows.
Nemotron
NVIDIA's Mamba-attention hybrid. A dense 4B, a 30B-A3B MoE, and the 3.5 Lightning 30B-A3B, each with a KV config for its few attention layers.
LFM2.5
Liquid AI's on-device series, 230M through 8B, plus a 3B that takes images. Convolutional blocks alternate with full attention, so only a few blocks hold a KV cache and it stays small as context grows.
Qwen
Three generations in one family: 3.8 with image input, 3.6, and 3.5, the daily driver behind Conjure and most of our agent work. 0.8B through 122B.
Gemma-4
Google's instruct and QAT series, e2b through 31B, with image input on the multimodal quants.
DeepSeek-V4
A 43-layer sparse-attention MoE that runs three different attention modes depending on the layer. The routed experts carry almost all the weight, so they go to 2-bit while attention stays higher.
Laguna
poolside's Laguna-XS-2.1, a sparse-MoE reasoning model for coding and agents. Its OptiQ quant nearly ties our best Capability Score.
Diffusion LLM
Our first non-autoregressive family. A discrete-diffusion Gemma that denoises a canvas of tokens instead of decoding left to right.
Mistral · Devstral
Devstral-Small-2, the tool-calling coding model. It was the first OptiQ quant of a Mistral-family base.
MiniCPM5
A strong sub-gigabyte 1B with a hybrid <think> reasoning mode. The smallest full-capability quant in the lineup.
Nanbeige
A looped transformer. It runs a 22-layer stack twice with shared weights, so it computes like a deeper network without the parameters.
Mage-VL
Microsoft's 5B vision-language model, the first video model in the lineup. It reads images and video, with the vision tower kept at bf16 in a sidecar.
And dozens more across the ecosystem, from us and from developers publishing their own quants. Browse every OptiQ quant on Hugging Face →
Standard MLX. Run mlx_lm.load("mlx-community/<repo>") and generate. Vendored-arch families (Laguna, Nanbeige, dhara) need import optiq first to register the arch.
Point optiq serve at any repo for an OpenAI and Anthropic-compatible endpoint with a mixed-precision KV cache.
Sensitivity-aware LoRA, hot-swappable adapters, and a full local Lab, all on your Mac.