Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
tags:
|
| 4 |
+
- hnpu
|
| 5 |
+
- hexagon
|
| 6 |
+
- npu
|
| 7 |
+
- v81
|
| 8 |
+
- text-generation
|
| 9 |
+
- 1-bit
|
| 10 |
+
base_model: prism-ml/Bonsai-27B
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# Bonsai-27B — 1-bit for the Qualcomm Hexagon NPU (HNPU)
|
| 14 |
+
|
| 15 |
+
A QHexRT-runnable bundle of [prism-ml/Bonsai-27B](https://huggingface.co/prism-ml/Bonsai-27B) at **true 1-bit
|
| 16 |
+
(Q1_0, ~1.125 bit/weight)** for the Hexagon NPU. Runs the full 27B model on-device via a custom HVX op-package.
|
| 17 |
+
|
| 18 |
+
**Supported arches:** v81 (SM8850).
|