2026年8月11日 技术热点总结

技术栈：TypeScript

核心介绍：直连腾讯 iLink，数据存 远端 Redis，登录用 LINUX DO OAuth。 Connects directly to Tencent iLink, stores data in remote Redis, and authenticates via LINUX DO OAuth. 功能 Features · 架构 Architecture · 快速开始 Quick Start · 文档 Docs · 许可证 License

项目数据：⭐ 690 Stars，🍴 505 Forks

项目简介：AI 短剧制作的 skill 集合：拆角色、出设定图、排大纲 | Agent skills for AI short-drama production — character bibles, model sheets, adaptation outlines. Runs in Claude Code & codex.

技术栈：JavaScript

核心介绍：> 我建了一个 AI 短剧交流群（付费），聊 AI 短剧的工作流、工具和实操。 > 有兴趣的加我：微信 hao_dev，添加时备注 github。 > 丢一本小说进去，出这个：

项目数据：⭐ 645 Stars，🍴 76 Forks

技术栈：Python

核心介绍：面向当前 ComfyUI 原生 MiniMax H3 的独立 T8 节点扩展。当前版本为 1.12.0，共注册 54 个节点，覆盖原生音画条件、对白边界分析、对白安全分轨混音、分时背景底轨锁定、来源视频音画重绘准备、音频控制与后处理、稳定双时钟采样、实验性多速率采样、 隔离的分段长视频续写、总时长编排、候选/接受状态与文件级合成、Ref2VA 单图/多图 参考的静态语义编辑，以及带异常释放保护、持久分段、精确时长后期和显式音色库的实验性语音链。 节点按稳定性与用途分为七个菜单： 本包不是把源音频简单塞进 latent：它按 ComfyUI 当前 H3 实现维护媒体展示顺序、 / / 标签、联合 AV latent、首尾关键帧、参考媒体和 噪声掩码之间的契约。

项目数据：⭐ 585 Stars，🍴 30 Forks

项目简介：The long-horizon computer-use harness. Run AI agents across desktop apps and the CLI for extended periods while preserving task state and making reliable progress on complex workflows. Features fresh-context execution, durable verified state, independent auditing, recoverable progress, and native Claude Code / Codex / OpenClaw integration.

技术栈：Python、agent、claude、claude-code、claude-plugin、cli、codex、codex-desktop、codex-plugin

核心介绍：Usage · What You Get · How It Works · Results · Project Website · 简体中文 > **The model determines what an agent can do in one round. LongHorizon-…

项目简介：MCP server for AI agents that need a phone number: order a private number in 200+ countries, read the SMS verification code, hand it back. The widest country coverage in the category, and you can check it with one API call.

技术栈：JavaScript、ai-agents、claude、claude-code、cline、cursor、mcp、mcp-server、model-context-protocol

核心介绍：Phone numbers as tools. Your agent orders a private number in the country a service expects, reads the SMS verification code straight from the API, and hands the number back. Nine tools over stdio, no SDK to learn. 200+ countries, the widest coverage in …

项目简介：ComfyUI MiniMax H3 Director workflow

核心介绍：这个仓库保存 5 份可直接导入 ComfyUI 的 MiniMax H3 Director 工作流，覆盖文生视频、图生视频、首尾帧、参考素材生视频、视频编辑和参考素材改视频。 工作流来自 AIMixer/ComfyUI_MiniMaxH3_Director。这里保留一套方便下载、复现和测试的副本。

项目数据：⭐ 468 Stars，🍴 50 Forks

World generative models are typically used through what they produce: a rendered future, a video-conditioned action, or latent context computed by a costly generative branch. We argue that their more reusable asset is the computation that constructs a future. As a generator transforms a corrupted future into a coherent trajectory, its intermediate states organize appearance, spatial layout, and interaction across levels of abstraction. Can this future-generative computation be internalized in a representation inferred from the present alone? We present Enfold, which transfers this computati…

Benchmarking video-language models has largely focused on short clips and single-sentence metrics, leaving open whether current systems can generate accurate long-form, paragraph-level descriptions. We introduce CLIP-CC-Bench, an evaluation suite for long-form video description built from 5 hours of movie content segmented into 90-second clips, each paired with an expert-written paragraph-style reference. The evaluation suite employs an ensemble of five state-of-the-art LLM-based embedding models to increase reliability and mitigate single-model bias, and applies two complementary methodolo…

Turn-taking is a central component of full-duplex interaction. Which turn-taking behaviors are appropriate varies with the scenario, yet current models apply a single norm regardless of context. This limitation originates in their training data: human-human speech corpora capture natural timing phenomena but provide little role grounding or scenario-specific norms, while heuristic or prompted synthesis methods inject turn-taking behaviors without basing them on human preferences. We introduce DuplexGen, a framework for generating dialogues with scenario-adaptive turn-taking by calibrating L…

Sequence models must decide what to write into memory and what to retain. In quantum and quantum-inspired sequence learning, nonlinear recurrent updates often require repeated circuit evaluations and sequential backpropagation through time, making long contexts costly. Gated fast-weight programmers (FWPs) based on quantum-inspired Kolmogorov-Arnold networks (QKANs) alleviate this bottleneck by storing context in time-varying fast parameters. However, their scalar gate applies one retention-write balance to every fast-state coordinate, forcing all parameters to share a memory timescale. We i…

Human evaluation of AI systems and digital products is costly, slow, and difficult to scale. Offline evaluations are more scalable but often abstract away human diversity and interactive behavior. We therefore introduce MatrAIx, a population-scale simulated-user evaluation infrastructure for testing AI systems and digital products with heterogeneous users. MatrAIx has three core components: First, Persona 8B contains 8.3 billion persona records represented by 1,290 categorical dimensions. Records are either sampled from a dependency graph that preserves correlated attributes or derived from…

CLI-based software-engineering agents have matured rapidly, yet the open ecosystem has converged on a single training environment: trajectory datasets used to fine-tune open models are collected almost exclusively under OpenHands. Models fine-tuned on this data score well under OpenHands but degrade substantially when deployed under any non-training scaffold. Untrained base models do not show this divergence, indicating the gap is fine-tuning-induced and tied to the conventions of the training scaffold. We argue that a load-bearing scaffold-specific behavior is planning structure, in two se…