johnsmith968530/WeiboAI-VibeThinker-1.5B-MLX-8bit

This model johnsmith968530/WeiboAI-VibeThinker-1.5B-MLX-8bit was converted to MLX format from WeiboAI/VibeThinker-1.5B using mlx-lm version 0.28.3.

export MODEL1_MAJOR="WeiboAI"
export MODEL1_MINOR="VibeThinker-1.5B"
export MODEL1_Q_BITS="8"
export MODEL1_SD="$(stardate)"

echodo () { echo "$@" && "$@"; }
mkdir -p /tmp/mlx_lm/convert

echodo time mlx_lm.convert \
  --hf-path "$MODEL1_MAJOR/$MODEL1_MINOR" \
  --mlx-path "/tmp/mlx_lm/convert/$MODEL1_SD" \
  --quantize \
  --q-bits "$MODEL1_Q_BITS" \
  --upload-repo \
    "johnsmith968530/$MODEL1_MAJOR-$MODEL1_MINOR-MLX-${MODEL1_Q_BITS}bit"
Downloads last month
74
Safetensors
Model size
0.4B params
Tensor type
BF16
·
U32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for johnsmith968530/WeiboAI-VibeThinker-1.5B-MLX-8bit

Base model

Qwen/Qwen2.5-1.5B
Quantized
(24)
this model