ToPo-ToPo/Qwen3.6-27B-mlx-bf16

Source: 🤗image-text-to-textapache-2.027.4B params55 GBsafetensors✓ 12 checksumsupdated 142d ago

MLX bf16 conversion of Qwen/Qwen3.6-27B for Apple Silicon (mlx-vlm).

Provenance (self-converted from official weights)

Source: Qwen/Qwen3.6-27B (license: apache-2.0)
Tool: mlx-vlm 0.6.3 — mlx_vlm.convert --hf-path Qwen/Qwen3.6-27B --mlx-path . --dtype bfloat16
Effective: 16.0 bits/weight
Validation: reproduced geometrically exact CAD output in an agentic CAD+FEM pipeline (volumes match the reference mlx-community conversion).

from mlx_vlm import load, generate
model, processor = load("ToPo-ToPo/Qwen3.6-27B-mlx-bf16")

CommentsShare your thoughts and shape the future of sovereign AI.

@ToPo-ToPo is unclaimedSign in with the matching Hugging Face account or an organization member account to verify this handle.Claim @ToPo-ToPo

Files and sources

SourceAvailable on 🤗

Checksums12 recorded

TorrentNot listed

SwarmNot checked

Weights

Original

Weights55 GB
11 files
model-00009-of-00011.safetensors 5.3 GBmodel-00005-of-00011.safetensors 5.3 GBmodel-00004-of-00011.safetensors 5.3 GBmodel-00006-of-00011.safetensors 5.3 GBmodel-00003-of-00011.safetensors 5.3 GBmodel-00007-of-00011.safetensors 5.3 GBmodel-00008-of-00011.safetensors 5.3 GBmodel-00002-of-00011.safetensors 5.3 GBmodel-00001-of-00011.safetensors 5.2 GBmodel-00010-of-00011.safetensors 4.3 GBmodel-00011-of-00011.safetensors 2.5 GB

Family

↑ Finetuned from Qwen/Qwen3.6-27B