Cloned from akmaldira/Qwen3-4B-Instruct-2507-RKLLM
Browse files- .gitattributes +1 -0
- README.md +51 -0
- qwen3-4b-instruct-2507_w8a8_rk3588.rkllm +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
qwen3-4b-instruct-2507_w8a8_rk3588.rkllm filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
library_name: rkllm
|
| 5 |
+
base_model: Qwen/Qwen3-4B-Instruct-2507
|
| 6 |
+
tags:
|
| 7 |
+
- rkllm
|
| 8 |
+
- rk3588
|
| 9 |
+
- npu
|
| 10 |
+
- rockchip
|
| 11 |
+
- quantized
|
| 12 |
+
- edge-ai
|
| 13 |
+
- orange-pi
|
| 14 |
+
model_name: Qwen3-4B-Instruct-2507-RKLLM
|
| 15 |
+
pipeline_tag: text-generation
|
| 16 |
+
|
| 17 |
+
---
|
| 18 |
+
|
| 19 |
+
# Qwen3-4B-Instruct-2507 for RK3588
|
| 20 |
+
|
| 21 |
+
This repository contains the optimized **RKLLM** (Language) models for running **Qwen3-4B-Instruct-2507** on Rockchip RK3588 NPUs.
|
| 22 |
+
|
| 23 |
+
### 📦 Files
|
| 24 |
+
|
| 25 |
+
| File | Type | Description |
|
| 26 |
+
| --- | --- | --- |
|
| 27 |
+
| `qwen3-4b-instruct-2507_w8a8_rk3588.rkllm` | **RKLLM** | Language Decoder (w8a8) |
|
| 28 |
+
|
| 29 |
+
---
|
| 30 |
+
|
| 31 |
+
### ⚙️ System Requirements
|
| 32 |
+
|
| 33 |
+
* **NPU Driver:** `≥ 0.9.8`
|
| 34 |
+
* **RKLLM Runtime:** `v1.2.3`
|
| 35 |
+
* **RKNN-Toolkit2:** `v2.3.0`
|
| 36 |
+
|
| 37 |
+
### 💻 Tested Hardware
|
| 38 |
+
|
| 39 |
+
* **Device:** Orange Pi 5 Pro
|
| 40 |
+
* **RAM:** 16GB
|
| 41 |
+
* **OS:** Armbian OS
|
| 42 |
+
|
| 43 |
+
---
|
| 44 |
+
|
| 45 |
+
### 🙏 Acknowledgements
|
| 46 |
+
|
| 47 |
+
* [Qwen Team](https://huggingface.co/qwen) for the base model architecture.
|
| 48 |
+
* [Rockchip / airockchip](https://github.com/airockchip/rknn-llm) for the RKLLM toolkit and runtime.
|
| 49 |
+
* Converted by [Akmal Dira](https://huggingface.co/akmaldira).
|
| 50 |
+
|
| 51 |
+
---
|
qwen3-4b-instruct-2507_w8a8_rk3588.rkllm
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:922e60a2f1b2989d6b486707b8b65fe6f923da097cffd388f09ddf63277efce7
|
| 3 |
+
size 4846781764
|