| --- |
| license: apache-2.0 |
| base_model: Qwen/Qwen3.6-27B |
| library_name: basert |
| pipeline_tag: text-generation |
| tags: |
| - basert |
| - apple-silicon |
| - quantized |
| --- |
| |
| # Qwen3.6-27B |
|
|
| [BaseRT](https://github.com/basecompute/baseRT) `.base` builds of [`Qwen/Qwen3.6-27B`](https://huggingface.co/Qwen/Qwen3.6-27B) for fast local inference on Apple Silicon (Metal). |
|
|
| A dense 27B reasoning model with hybrid attention (Gated DeltaNet + periodic full attention). Q4 needs ~24 GB Apple Silicon (tight); Q8 needs 48 GB+. |
|
|
| ## Files |
|
|
| | File | Precision | Size | |
| |------|-----------|------| |
| | `Qwen3.6-27B-Q4.base` | 4-bit | 15 GB | |
| | `Qwen3.6-27B-Q8.base` | 8-bit | 27 GB | |
|
|
| ## Usage |
|
|
| ```bash |
| curl -LsSf https://basecompute.co/install.sh | sh |
| basert pull basecompute/Qwen3.6-27B |
| basert chat basecompute/Qwen3.6-27B |
| ``` |
|
|
| Released under the apache-2.0 license, inherited from the base model. |
|
|
