opensource/FireRed-OpenStoryline
3
0
Fork 0
mirror of https://github.com/FireRedTeam/FireRed-OpenStoryline.git synced 2026-09-17 16:42:32 +08:00
Code Issues Packages Projects Releases Wiki Activity

refactor: remove developer chat model config (#52)

Browse Source 
Co-authored-by: 杨其伟 <yangqiwei1@xiaohongshu.com>
This commit is contained in:
Anlittledy
2026-03-11 15:49:20 +08:00
committed by GitHub
co-authored by 杨其伟
parent 8eb45a06e6
commit 1229037f19
6 changed files with 97 additions and 109 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/source/en/api-key.md
+12 -11
View File
@@ -21,8 +21,12 @@ Note: For users outside China, we recommend using large language models such as
- **API Key**: Fill in the Key obtained in the previous step
3. **API Configuration**
- **Web Usage**: Select "Use Custom Model" in the LLM model form, and fill in the model according to the configuration parameters
- **Local Deployment**: In config.toml, locate `[developer.chat_models_config."deepseek-chat"]` and fill in the configuration parameters to make the default configuration accessible from the Web page. Locate `[llm]` and configure model, base_url, and api_key
- **Web Usage**:
- In the LLM model dropdown, select **Custom Model**, then fill in the model settings according to your configuration parameters.
- Or, open `config.toml`, locate `[llm]`, and configure `model`, `base_url`, and `api_key`. The model you entered will then appear in the dropdown on the Web page.
- **CLI**:
- If you prefer the CLI entry point, you need to open `config.toml`, locate `[llm]`, and configure `model`, `base_url`, and `api_key`.
## 2. Multimodal Large Language Model (VLM)
@@ -42,15 +46,12 @@ Note: For users outside China, we recommend using large language models such as
- **Model Name**: `qwen3-vl-8b-instruct`
- **Base URL**: `https://dashscope.aliyuncs.com/compatible-mode/v1`
- Parameter Configuration: Select "Use Custom Model" in the VLM Model form and fill in the parameters. For local deployment, locate `[vlm]` and configure model, base_url, and api_key. Add the following fields in config.toml as the default Web API configuration:
```
[developer.chat_models_config."qwen3-vl-8b-instruct"]
base_url = "https://dashscope.aliyuncs.com/compatible-mode/v1"
api_key = "YOUR_API_KEY"
timeout = 20.0
temperature = 0.1
max_retries = 2
```
- Parameter Configuration:
- **Web Usage**:
- In the VLM model dropdown, select **Custom Model**, then fill in the model settings according to your configuration parameters.
- Or, open `config.toml`, locate `[vlm]`, and configure `model`, `base_url`, and `api_key`. The model you entered will then appear in the dropdown on the Web page.
- **CLI**:
- If you prefer the CLI entry point, you need to open `config.toml`, locate `[vlm]`, and configure `model`, `base_url`, and `api_key`.
### 2.3 Using Qwen3-Omni
docs/source/zh/api-key.md
+11 -11
View File
@@ -21,8 +21,11 @@
- **API Key**：填写上一步获取的 Key
3. **API填写**
- **Web使用**: 在LLM模型表单中选择使用自定义模型，模型按照配置参数进行填写
- **本地部署**: 在config.toml中 找到`[developer.chat_models_config."deepseek-chat"]` 将配置参数填写上去，使得Web页面可以访问到该默认配置。 找到`[llm]`并配置model、base_url、api_key
- **Web使用**:
- 在LLM模型下拉框中选择使用自定义模型，模型按照配置参数进行填写
- 或是在`config.toml`中 找到`[llm]`并配置model、base_url、api_key。Web页面下拉框会出现你填写的模型。
- **CLI**：
- 如果你偏好 CLI 入口，需要在`config.toml`中找到`[llm]`并配置model、base_url、api_key。
## 二、多模态大模型 (VLM)
@@ -42,15 +45,12 @@
- **模型名称**：`qwen3-vl-8b-instruct`
- **Base URL**：`https://dashscope.aliyuncs.com/compatible-mode/v1`
- **参数填写**：在VLM Model表单中选择"使用自定义模型"进行参数填写。本地部署时，找到`[vlm]`并配置model、base_url、api_key，在config.toml中新增以下字段作为Web的API默认配置：
```
[developer.chat_models_config."qwen3-vl-8b-instruct"]
base_url = "https://dashscope.aliyuncs.com/compatible-mode/v1"
api_key = "YOUR_API_KEY"
timeout = 20.0
temperature = 0.1
max_retries = 2
```
- **参数填写**：
- **Web使用**:
- 在VLM模型下拉框中选择使用自定义模型，模型按照配置参数进行填写。
- 或是在`config.toml`中 找到`[vlm]`并配置model、base_url、api_key。Web页面下拉框会出现你填写的模型。
- **CLI**：
- 如果你偏好 CLI 入口，需要在`config.toml`中找到`[vlm]`并配置model、base_url、api_key。
### 2.3 使用Qwen3-Omni