Update README.md
Browse files
README.md
CHANGED
@@ -12,86 +12,29 @@ tags:
|
|
12 |
license: other
|
13 |
---
|
14 |
|
15 |
-
#
|
16 |
|
17 |
-
- [中文](#通义千问-1_8b)
|
18 |
- [English](#qwen-1_8b-chat_rkllm)
|
19 |
-
|
20 |
-
## 通义千问-1_8B
|
21 |
-
|
22 |
-
### 介绍
|
23 |
-
|
24 |
-
通义千问-1_8B 是从 [Qwen-1_8B](https://huggingface.co/Qwen/Qwen-1_8B/) 转换而来的 RKLLM 模型,专为 Rockchip 设备优化。该模型运行于 RK3588 的 NPU 上。
|
25 |
-
|
26 |
-
- **模型名称**: Qwen-1_8B
|
27 |
-
- **模型架构**: 与 Qwen-1_8B 相同
|
28 |
-
- **发布者**: FydeOS
|
29 |
-
- **发布日期**: 2024-05-14
|
30 |
-
|
31 |
-
### 模型详情
|
32 |
-
|
33 |
-
通义千问-1_8B(Qwen-1_8B)是阿里云研发的通义千问系列大模型中的一员,拥有 18 亿参数。该模型基于 Transformer 架构,经过大规模预训练,涵盖了多种类型的数据,包括网络文本、专业书籍和代码。
|
34 |
-
|
35 |
-
### 使用指南
|
36 |
-
|
37 |
-
> 此模型仅支持搭载 Rockchip RK3588/s 芯片的设备。请确认设备信息并确保 NPU 可用。
|
38 |
-
|
39 |
-
#### openFyde 系统
|
40 |
-
|
41 |
-
> 请确保你已将系统升级到最新版本。
|
42 |
-
|
43 |
-
1. 下载模型文件 `XXX.rkllm`。
|
44 |
-
2. 新建文件夹 `model/`,将模型文件放置于该文件夹内。
|
45 |
-
3. 启动 FydeOS AI,在设置页面进行相关配置。
|
46 |
-
|
47 |
-
#### 其它系统
|
48 |
-
|
49 |
-
> 请确保已完成 RKLLM 的 NPU 相关内核更新。
|
50 |
-
|
51 |
-
1. 下载模型文件 `XXX.rkllm`。
|
52 |
-
2. 按照官方文档进行配置:[官方文档](https://github.com/airockchip/rknn-llm)。
|
53 |
-
|
54 |
-
### 常见问题(FAQ)
|
55 |
-
|
56 |
-
如遇到问题,请先查阅 issue 区,若问题仍未解决,再提交新的 issue。
|
57 |
-
|
58 |
-
### 限制与注意事项
|
59 |
-
|
60 |
-
- 模型在某些情况下可能存在性能限制
|
61 |
-
- 使用时请遵循相关法律法规
|
62 |
-
- 可能需要进行适当的参数调优以达到最佳效果
|
63 |
-
|
64 |
-
### 许可证
|
65 |
-
|
66 |
-
本模型采用与 [Qwen-1_8B](https://huggingface.co/Qwen/Qwen-1_8B/) 相同的许可证。
|
67 |
-
|
68 |
-
### 联系方式
|
69 |
-
|
70 |
-
如需更多信息,请联系:
|
71 |
-
|
72 |
-
- **电子邮件**: hi@fydeos.io
|
73 |
-
- **主页**: [FydeOS AI](https://fydeos.ai/zh/)
|
74 |
-
|
75 |
-
|
76 |
|
77 |
## Qwen-1_8B-Chat_rkLLM
|
78 |
|
79 |
### Introduction
|
80 |
|
81 |
-
Qwen-1_8B-Chat_rkLLM is a model converted from [Qwen-1_8B](https://huggingface.co/Qwen/Qwen-1_8B/),
|
82 |
|
83 |
- **Model Name**: Qwen-1_8B
|
84 |
-
- **Architecture**:
|
85 |
- **Publisher**: FydeOS
|
86 |
-
- **Release Date**: 14
|
87 |
|
88 |
### Model Details
|
89 |
|
90 |
-
Qwen-1_8B-Chat_rkLLM is part of the Tongyi Qianwen series developed by Alibaba
|
91 |
|
92 |
### User Guide
|
93 |
|
94 |
-
This model is only
|
95 |
|
96 |
#### openFyde
|
97 |
|
@@ -101,16 +44,12 @@ This model is only supported on devices with the Rockchip RK3588/s chip. Please
|
|
101 |
2. Create a folder named `model/` and place the model file inside this folder.
|
102 |
3. Launch FydeOS AI and configure the settings on the settings page.
|
103 |
|
104 |
-
#### Other
|
105 |
|
106 |
> Ensure you have updated the NPU kernel related to RKLLM.
|
107 |
|
108 |
1. Download the model file `XXX.rkllm`.
|
109 |
-
2. Follow the configuration guidelines
|
110 |
-
|
111 |
-
### FAQ
|
112 |
-
|
113 |
-
For issues, please refer to the issue section. If your problem remains unresolved, submit a new issue.
|
114 |
|
115 |
### Limitations and Considerations
|
116 |
|
@@ -118,13 +57,42 @@ For issues, please refer to the issue section. If your problem remains unresolve
|
|
118 |
- Ensure compliance with relevant laws and regulations during usage.
|
119 |
- Parameter tuning might be necessary to achieve optimal performance.
|
120 |
|
121 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
122 |
|
123 |
-
|
124 |
|
125 |
-
|
126 |
|
127 |
-
|
128 |
|
129 |
-
|
130 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
license: other
|
13 |
---
|
14 |
|
15 |
+
# Qwen-1_8B-Chat_rkLLM
|
16 |
|
|
|
17 |
- [English](#qwen-1_8b-chat_rkllm)
|
18 |
+
- [中文](#通义千问-1_8b)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
19 |
|
20 |
## Qwen-1_8B-Chat_rkLLM
|
21 |
|
22 |
### Introduction
|
23 |
|
24 |
+
Qwen-1_8B-Chat_rkLLM is a model converted from [Qwen-1_8B](https://huggingface.co/Qwen/Qwen-1_8B/), optimized for Rockchip devices. This model runs on the NPU of the RK3588 chip.
|
25 |
|
26 |
- **Model Name**: Qwen-1_8B
|
27 |
+
- **Architecture**: Same as Qwen-1_8B
|
28 |
- **Publisher**: FydeOS
|
29 |
+
- **Release Date**: May 14, 2024
|
30 |
|
31 |
### Model Details
|
32 |
|
33 |
+
Qwen-1_8B-Chat_rkLLM is part of the Tongyi Qianwen series developed by Alibaba. This large language model features 1.8 billion parameters and is based on the Transformer architecture. It has been pre-trained on a diverse dataset, including web texts, professional literature, and code.
|
34 |
|
35 |
### User Guide
|
36 |
|
37 |
+
This model is supported only on devices with the Rockchip RK3588/s chip. Please verify your device’s chip information and ensure the NPU is operational.
|
38 |
|
39 |
#### openFyde
|
40 |
|
|
|
44 |
2. Create a folder named `model/` and place the model file inside this folder.
|
45 |
3. Launch FydeOS AI and configure the settings on the settings page.
|
46 |
|
47 |
+
#### Other OS
|
48 |
|
49 |
> Ensure you have updated the NPU kernel related to RKLLM.
|
50 |
|
51 |
1. Download the model file `XXX.rkllm`.
|
52 |
+
2. Follow the configuration guidelines in the [official documentation](https://github.com/airockchip/rknn-llm).
|
|
|
|
|
|
|
|
|
53 |
|
54 |
### Limitations and Considerations
|
55 |
|
|
|
57 |
- Ensure compliance with relevant laws and regulations during usage.
|
58 |
- Parameter tuning might be necessary to achieve optimal performance.
|
59 |
|
60 |
+
## 通义千问-1_8B
|
61 |
+
|
62 |
+
### 介绍
|
63 |
+
|
64 |
+
通义千问-1_8B 是从 [Qwen-1_8B](https://huggingface.co/Qwen/Qwen-1_8B/) 转换而来的 RKLLM 模型,专为 Rockchip 设备优化。该模型运行于 RK3588 的 NPU 上。
|
65 |
+
|
66 |
+
- **模型名称**: Qwen-1_8B
|
67 |
+
- **模型架构**: 与 Qwen-1_8B 相同
|
68 |
+
- **发布者**: FydeOS
|
69 |
+
- **发布日期**: 2024-05-14
|
70 |
+
|
71 |
+
### 模型详情
|
72 |
+
|
73 |
+
通义千问-1_8B(Qwen-1_8B)是阿里云研发的通义千问系列大模型中���一员,拥有 18 亿参数。该模型基于 Transformer 架构,经过大规模预训练,涵盖了多种类型的数据,包括网络文本、专业书籍和代码。
|
74 |
+
|
75 |
+
### 使用指南
|
76 |
|
77 |
+
此模型仅支持搭载 Rockchip RK3588/s 芯片的设备。请确认设备信息并确保 NPU 可用。
|
78 |
|
79 |
+
#### openFyde 系统
|
80 |
|
81 |
+
> 请确保你已将系统升级到最新版本。
|
82 |
|
83 |
+
1. 下载模型文件 `XXX.rkllm`。
|
84 |
+
2. 新建文件夹 `model/`,将模型文件放置于该文件夹内。
|
85 |
+
3. 启动 FydeOS AI,在设置页面进行相关配置。
|
86 |
+
|
87 |
+
#### 其它系统
|
88 |
+
|
89 |
+
> 请确保已完成 RKLLM 的 NPU 相关内核更新。
|
90 |
+
|
91 |
+
1. 下载模型文件 `XXX.rkllm`。
|
92 |
+
2. 按照官方文档进行配置:[官方文档](https://github.com/airockchip/rknn-llm)。
|
93 |
+
|
94 |
+
### 限制与注意事项
|
95 |
+
|
96 |
+
- 模型在某些情况下可能存在性能限制
|
97 |
+
- 使用时请遵循相关法律法规
|
98 |
+
- 可能需要进行适当的参数调优以达到最佳效果
|