Downloads · 30 days
1.1K
21% of all-time downloads
khanh2023/Qwen3.6-14B-A3B-FableVibes-mlx-q4
Qwen3.6-14B-A3B-FableVibes-mlx-q4 is a text generation model from khanh2023. Use it when you need the model to write or continue text. It is set up for mlx. The card lists the license as apache-2.0.
MLX 4-bit quantization of tvall43/Qwen3.6-14B-A3B-FableVibes, for local inference on Apple Silicon.
Downloads · 30 days
1.1K
21% of all-time downloads
All-time downloads
5.5K
Public
Parameters
13.8B
7.8 GB on disk
Likes
6
Public
Click a slice to open those files.
.safetensors7.7 GB · 100%
How the weights are stored.
U3213.8B · 100%
From the Hugging Face model README
MLX 4-bit quantization of tvall43/Qwen3.6-14B-A3B-FableVibes, for local inference on Apple Silicon.
This repo is only a quantized MLX conversion. All credit for the model itself goes to the original author, tvall43. Please see and cite the original model card.
The base is a REAP-pruned Qwen3.6-35B-A3B reduced to ~14B total / ~3B active (90 experts, 8 active), recovered with a QLoRA distill of Claude Fable 5 reasoning traces. It uses the Qwen3.5 hybrid architecture (GatedDeltaNet linear attention + full attention + MoE) and emits <think>...</think> reasoning.
experts.{i}.{gate,up,down}_proj) into mlx-lm's stacked experts.gate_up_proj / experts.down_proj format.mlx-lm.uv run python -m mlx_lm generate \
--model khanh2023/Qwen3.6-14B-A3B-FableVibes-mlx-q4 \
--prompt "Solve: ..."
-mlx-q6 variant is also available.