mlx-community/Groot-19B-A3B-OptiQ-4bit

Built with mlx-optiq, the MLX-native toolkit to quantize, fine-tune, and serve LLMs locally on Apple Silicon. All OptiQ models · Docs

A 4-bit mixed-precision MLX quant of GrootTeam/Groot-19B-A3B-261023, a reasoning model post-trained from Qwen3.6-35B-A3B. It has 128 routed experts per layer and activates 8 per token. Text, tool calling and image input all work.

On disk	13.2 GB, including the 0.9 GB vision tower (source BF16 38.9 GB)
Parameters	19B total, about 3B active per token
Precision	118 tensors at 4-bit, 394 at 8-bit, vision tower at bf16

How it was made

The per-layer bit allocation is reused from Qwen3.6-35B-A3B-OptiQ-4bit-REAP-19B, which has the same 128-expert layout. All 512 tensors matched. No sensitivity sweep was run on this checkpoint.

A full Capability Score was not run before publishing. Before upload, it was served with optiq serve and checked on facts, arithmetic, code, a tool call and an image.

The MTP head from the source is not included. It does not attach in OptiQ yet, so it would add 0.8 GB with no benefit.

Use it

pip install mlx-optiq
optiq serve --model mlx-community/Groot-19B-A3B-OptiQ-4bit

from mlx_lm import load, generate
model, tok = load("mlx-community/Groot-19B-A3B-OptiQ-4bit")
print(generate(model, tok, prompt="Hello", max_tokens=64))