GeekAI-PLUS-v4.1.8

418 is a major version update. We have upgraded the UI, added theme switching support, integrated the Gitee AI API, enabled OpenAI real-time voice call functionality, and added Markdown emoji parsing.

Update Details

Feature Optimization: Completely revamped UI with theme switching support. 🚀 🚀 🚀
New Feature: Integrated Gitee AI API, currently supporting Gitee's SD drawing interface and AI chat interface. 🚀 🚀 🚀
Bug Fix: Fixed an issue where the Luma API update caused task response parsing to fail.
Feature Optimization: Added support for Suno v4.0 model.
Bug Fix: Fixed an issue where completed Suno tasks could not be deleted.
New Feature: Added OpenAI real-time voice call functionality, now supporting pay-per-use and allowing administrators to configure computing resource consumption per call.
New Feature: Generating prompts now consumes computing resources. Administrators can configure the resource consumption per prompt generation to prevent abuse.
New Feature: DALL-E-3 drawing now supports the Flux drawing model. Administrators can add Flux, SD, and other drawing models in the backend.
Feature Optimization: Markdown now supports emoji parsing.
Feature Optimization: When a drawing menu is disabled in the admin backend, the corresponding mobile drawing menu will also be disabled (the feature will not be displayed).

1. Update Database Structure

ALTER TABLE `chatgpt_suno_jobs` CHANGE `err_msg` `err_msg` VARCHAR(1024) CHARACTER SET utf8mb4 COLLATE utf8mb4_0900_ai_ci NULL DEFAULT NULL COMMENT '错误信息';

ALTER TABLE `chatgpt_sd_jobs` CHANGE `err_msg` `err_msg` VARCHAR(1024) CHARACTER SET utf8mb4 COLLATE utf8mb4_0900_ai_ci NULL DEFAULT NULL COMMENT '错误信息';

ALTER TABLE `chatgpt_mj_jobs` CHANGE `err_msg` `err_msg` VARCHAR(1024) CHARACTER SET utf8mb4 COLLATE utf8mb4_0900_ai_ci NULL DEFAULT NULL COMMENT '错误信息';

ALTER TABLE `chatgpt_dall_jobs` CHANGE `err_msg` `err_msg` VARCHAR(1024) CHARACTER SET utf8mb4 COLLATE utf8mb4_0900_ai_ci NOT NULL COMMENT '错误信息';
ALTER TABLE `chatgpt_video_jobs` CHANGE `err_msg` `err_msg` VARCHAR(1024) CHARACTER SET utf8mb4 COLLATE utf8mb4_0900_ai_ci NULL DEFAULT NULL COMMENT '错误信息';
ALTER TABLE `chatgpt_chat_models` ADD `type` VARCHAR(10) NOT NULL DEFAULT 'chat' COMMENT '模型类型 (chat, img)' AFTER `id`;

2. Update Image Version

Edit docker-compose.yaml to update the image versions for geekai-api and geekai-web.

# Backend API Service
geekai-api:
  image: registry.cn-shenzhen.aliyuncs.com/geekmaster/geekai-plus-api:v4.1.8-amd64
  container_name: geekai-api
  restart: always
  ports:
    - '5678:5678'
    - '6012:6012'
  volumes:
    - /usr/share/zoneinfo/Asia/Shanghai:/etc/localtime
    - ./conf/config.toml:/var/www/app/config.toml
    - ./logs:/var/www/app/logs
    - ./static:/var/www/app/static
    - ./res/sd:/var/www/app/res/sd
    - ./certs:/var/www/app/certs
    - ./data:/var/www/app/data
  environment:
    - APP_DEBUG=false
    - LOG_LEVEL=info
    - CONFIG_FILE=config.toml

# Frontend Application
geekai-web:
  image: registry.cn-shenzhen.aliyuncs.com/geekmaster/geekai-plus-web:v4.1.8-amd64
  container_name: geekai-web
  restart: always
  ports:
    - '9004:8080'
  volumes:
    - ./logs/nginx:/var/log/nginx
    - ./conf/nginx/conf.d:/etc/nginx/conf.d
    - ./conf/nginx/nginx.conf:/etc/nginx/nginx.conf
    - ./ssl:/etc/nginx/ssl

3. Restart Containers

docker-compose down
docker-compose up -d

4. Model Categorization

418 introduces model categorization in the admin backend, distinguishing between chat models and drawing models.

Model Categorization

We have integrated the DALL-E drawing page, supporting not only OpenAI's DALL-E-3 but also Flux, SD, and other AI drawing tools compatible with the OpenAI Image API.

DALL-E Drawing Page

Simply add a "Drawing" model in the admin backend, and it will appear in the frontend drawing page.

Drawing Model

5. Support for OpenAI Advanced Voice Model

In version 416, we introduced OpenAI's real-time voice call feature, but many users reported issues with cost control. Therefore, version 418 supports OpenAI's advanced voice model, enabling pay-per-use and allowing administrators to configure computing resource consumption per call.

Step 1: In the admin backend, navigate to "System Settings" > "Resource Configuration" > "Advanced Voice Resource" to configure the computing resource consumption per real-time voice call.

Advanced Voice Resource Configuration

Step 2: In the admin backend, go to "API-KEY" and add a "Realtime" API KEY.

Add Realtime API KEY

Step 3: On the frontend chat page, click the "Advanced Voice" button to start a real-time voice call.

Advanced Voice Button

Once the connection is established, a voice call window will pop up, allowing users to engage in real-time voice conversations.

Calls are charged per use, with no time limit per call. You can chat as long as you want until the conversation ends abnormally or the user hangs up.

Voice Call Window

6. Support for Gitee AI API

GeekAI version 418 now seamlessly integrates the Gitee AI API! Gitee AI's Serverless API provides enterprise-grade large model API services out of the box.

✨ High Cost-Effectiveness - Compared to other platforms, Gitee AI API offers more affordable pricing, significantly reducing your AI development costs.
🚀 Lightning-Fast Response - With major domestic AI computing centers onboard, the API responds as fast as lightning, ensuring a smoother application experience.
🎯 Full Compatibility - Supports mainstream open-source models like Qwen2.5-72B, ChatGLM, Stable Diffusion, Flux, ChatTTS, and CosyVoice.
🎁 Daily Free Quota - Enjoy 100 free API calls per day for testing and development.
💪 Stable and Reliable - Backed by the Gitee platform, ensuring service stability.

For instructions on integrating the Gitee AI API, refer to the Gitee AI API Integration Guide.