Repo	GGUF Equivalent	Size
Brooooooklyn/Qwen3.5-27B-UD-Q2_K_XL-mlx	UD-Q2_K_XL	15 GB
Brooooooklyn/Qwen3.5-27B-UD-Q3_K_XL-mlx	UD-Q3_K_XL	17 GB
Brooooooklyn/Qwen3.5-27B-UD-Q4_K_XL-mlx	UD-Q4_K_XL	20 GB
Brooooooklyn/Qwen3.5-27B-UD-Q5_K_XL-mlx	UD-Q5_K_XL	24 GB
Brooooooklyn/Qwen3.5-27B-UD-Q6_K_XL-mlx	UD-Q6_K_XL	27 GB
Brooooooklyn/Qwen3.5-27B-UD-Q8_K_XL-mlx	UD-Q8_K_XL	29 GB

Per-Tensor Bit Assignments (N=4)

Weight	Bits	Rationale
`embed_tokens`	6-bit	KLD ~0.15 — very low sensitivity
`lm_head`	8-bit	KLD ~0.05 — safest tensor
`self_attn.q/k/v_proj`	6-bit + AWQ	KLD ~1.5-2.9, AWQ via layernorm
`linear_attn.in_proj_qkv/z`	6-bit + AWQ	KLD ~2.9, AWQ via layernorm
`self_attn.o_proj`	bf16	NOT AWQ-correctable
`linear_attn.out_proj`	bf16	KLD ~6.0 — worst tensor
`down_proj`	5-bit	"Slightly more sensitive"
`gate_proj`, `up_proj`	4-bit	"Generally ok" at low bits

Based on Unsloth Dynamic 2.0 per-tensor KLD analysis with imatrix AWQ pre-scaling.

Usage

typescript

import {{ loadModel }} from '@mlx-node/lm';
const model = await loadModel('./Qwen3.5-27B-UD-Q4_K_XL-mlx');
const result = await model.chat(
  [{{ role: 'user', content: 'Hello!' }}],
  {{ maxNewTokens: 2048, temperature: 0.6, enableThinking: false }},
);
console.log(result.text);

How It Was Made

mlx convert -i Qwen3.5-27B -o Qwen3.5-27B-UD-Q4_K_XL-mlx -q --q-bits 4 --q-recipe unsloth --imatrix-path imatrix_unsloth.gguf

Acknowledgments

Unsloth — Per-layer KLD benchmarks and Dynamic 2.0 methodology
Qwen Team — Qwen3.5 model family
Apple MLX — Metal-accelerated ML framework

License

Apache 2.0 (inherited from base model).

Brooooooklyn/Qwen3.5-27B-UD-Q4_K_XL-mlx

Qwen3.5-27B — UD-Q4_K_XL (mlx-node)

All Variants

Per-Tensor Bit Assignments (N=4)

Usage

How It Was Made

Acknowledgments

License

Brooooooklyn/
Qwen3.5-27B-UD-Q4_K_XL-mlx