Downloads · 30 days
43
7% of all-time downloads
Qwen3.6-27B-OBLITERATED-MLX-4bit
edward-lcl/Qwen3.6-27B-OBLITERATED-MLX-4bit
Qwen3.6-27B-OBLITERATED-MLX-4bit is a text generation model from edward-lcl. 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 conversion of OBLITERATUS/Qwen3.6-27B-OBLITERATED for Apple Silicon.
All-time downloads
583
Public
Parameters
26.9B
15.2 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors15.1 GB · 100%
Parameter types
How the weights are stored.
U3226.9B · 100%
Try a prompt
Base models
Model card
From the Hugging Face model README
Qwen3.6-27B-OBLITERATED — MLX 4-bit
MLX conversion of OBLITERATUS/Qwen3.6-27B-OBLITERATED for Apple Silicon.
Model details
| Base | Qwen3.6-27B-OBLITERATED |
| Architecture | qwen3_5 (mixed linear + full attention) |
| Parameters | 26.9B |
| Quantization | 4-bit affine, group_size 64 |
| Format | MLX safetensors (3 shards, ~14 GB) |
| Context window | 262,144 tokens |
| Tokenizer | Qwen2Tokenizer with tool-call support |
The model uses an alternating linear/full attention pattern across 64 layers (mamba-style linear attention interleaved with standard multi-head attention), giving it efficient long-context performance on Apple Silicon.
Usage
mlx-lm
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("edward-lcl/Qwen3.6-27B-OBLITERATED-MLX-4bit")
response = generate(model, tokenizer, prompt="Your prompt here", max_tokens=512)
print(response)
mlx-lm CLI
mlx_lm.generate \
--model edward-lcl/Qwen3.6-27B-OBLITERATED-MLX-4bit \
--prompt "Your prompt here" \
--max-tokens 512
Generation defaults
| Temperature | 0.35 |
| Repetition penalty | 1.05 |
| top_p | 1.0 |
Notes
- Converted and tested on Apple M5 Pro, 48 GB unified memory
- Metal's effective GPU working set ceiling is ~37.4 GB regardless of total RAM; model weights occupy ~14.4 GB of that
- Practical context on 48 GB: comfortable sustained use up to ~75K tokens; hard ceiling observed at ~91K before prefill guard trips
- Source model includes a
chat_template.jinjawith full tool-call formatting
Source
Original model: OBLITERATUS/Qwen3.6-27B-OBLITERATED
- Task
- Text Generation
- Library
- mlx
- Type
- qwen3_5
- License
- apache-2.0
- Languages
- en
- Created
- Jun 9, 2026
- Updated
- Jun 9, 2026