 

AI打造唐朝历史互动体验应用

2025-12-19 分类：前沿哨所阅读(2) 评论(0) 赞(0)

智谱 GLM，支持多语言、多任务推理。从写作到代码生成，从搜索到知识问答，AI 生产力的中国解法。

这款应用基于谷歌AI技术，打造了一个沉浸式的唐朝历史体验平台。用户可以通过左侧对话框与不同角色互动，包括西市胡商、平康坊乐人、金吾卫小吏和落第书生，并切换到贞观、开元、元和、大顺等唐朝时期。右上角提供今日简报功能，帮助用户了解历史背景。该应用展示了AI在历史教育和文化传承中的创新应用，为科技爱好者提供了探索古代文明的互动方式。体验地址已提供，欢迎用户尝试。

原文链接：Linux.do

赞(0)

未经允许不得转载：Toy Tech Blog » AI打造唐朝历史互动体验应用

分享到

AI 唐朝历史谷歌

评论抢沙发

快讯

Eino Framework's Dual-Path Support: Is the AI Agent Facing an Identity Crisis?

The Go-based LLM framework Eino supports both LangChain and LangGraph-based workflow orchestration, as well as Google ADK-like agent operations. These two agent implementations have different principles, raising concerns about the framework's positioning: due to the overly fragmented APIs, Eino may struggle to stand out in specific domains and eventually be phased out by the market. This discussion reflects the challenge of balancing flexibility and focus in AI framework design, offering valuable insights for developers choosing tools. While Eino's innovative attempt is commendable, community skepticism about its sustainability reminds the industry that frameworks need clear core advantages to avoid becoming "jack-of-all-trades, master of none" casualties.

Original Link:Linux.do

49分钟前
Claude Code API Error 400 Fix: Custom Commands to Bypass Thinking Block Limitations

Suddenly encountering API Error: 400 while using Claude Code, making conversations unusable? Even the built-in /resume command fails to restore functionality. This article dives deep into the root cause: Claude's extended thinking mode generates unmodifiable thinking blocks, and when subsequent requests attempt to modify these blocks, a 400 error is triggered. The article provides two solutions: first, creating a custom /resume-new command that bypasses the issue by reading only conversation summaries instead of complete history; second, restoring conversations after disabling thinking mode. The /resume-new command is more elegant, as it doesn't inherit problematic history and can proactively provide context introduction. The article details the creation method, execution flow, and differences from the built-in /resume, offering a practical guide for developers facing the same issue.

Original link:Linux.do

49分钟前
Codex Review Command Error: Invalid ID Format Issue

On the Linux.do technical forum, a user reported a specific bug encountered while using the Codex AI model. The issue is described as follows: within the same session, when a user first uses the /review command to get review results and then attempts to have the model modify the content, the system returns an error. The error details are: 'Invalid 'input[2].id': 'review:rollout:user'. Expected an ID that contains letters, numbers, underscores, or dashes, but this value contained additional characters.' This indicates that the ID 'review:rollout:user' contains illegal characters such as colons, violating the ID format specification. The user is asking how to resolve this issue. This bug involves the API implementation and ID validation logic of the Codex model, and for AI developers, researchers, and users of Codex, it is a technical issue worth paying attention to, as it may affect the reliability of the model in automated tasks.

Original Link:Linux.do

49分钟前
GPT-5.2-Codex vs Claude-Sonnet-4.5: Code Generation Showdown

In the Linux.do community, a user conducted an in-depth comparison of GPT-5.2-Codex and Claude-Sonnet-4.5's performance in the code generation field. The user asked which model excels at writing code and shared hands-on experience: Codex frequently fails after five retry attempts, despite network latency being kept under 100ms and a clean environment. The discussion focuses on the practicality and stability of AI models in software engineering, providing valuable real-world insights for developers and revealing the challenges and advantages of cutting-edge AI tools in actual deployment.

Original article:Linux.do

50分钟前
AI Text Adventure Game 'Floating Life Ten Dreams' Releases Docker Deployment Version

'Floating Life Ten Dreams' is a web-based immersive text adventure game where players take on the role of characters battling fate. With ten daily opportunities to enter different dreams, players experience unique life stories dynamically generated by AI. The game's core focuses on the choice mechanism between 'contentment' and 'greed.' To solve environment configuration challenges, the developer has created a Docker containerized deployment solution, recommending the use of docker-compose to simplify the installation process. This open-source project supports L-site login and a redeem code generation toggle, with a pure-play mode publicly available by default. The project has been released on GitHub, where users can report issues or contribute code. This solution highlights the innovative application of AI technology in game narrative and Docker's technical value in simplifying deployment, providing developers with a practical tool.

Original Link:Linux.do

50分钟前
DIY AI Role-Playing Prompt Marketplace Launches with Gemini Model Support

JiuPawMiao is a self-built AI role-playing prompt marketplace, currently featuring 120 pieces of content, with the website still under development. The featured section offers general templates and general limit-breaking templates, which have been tested and confirmed to support the Gemini-3-Pro model. To ensure website stability, R18 content requires login for viewing, and users can quickly register via GitHub. This project is maintained by an individual developer, and user feedback is encouraged to improve the experience. As an AI interaction resource platform, it provides a convenient channel for role-playing enthusiasts to obtain prompts, promotes the application of AI in the field of role-playing, and offers practical resources for developers and enthusiasts.

Original Link:Linux.do

51分钟前