1
又快又聪明，阿里发布Qwen-Audio-3.0-Realtime：实时语音大模型四项功能升级
阿里实时语音大模型四项功能同步升级，智商与口语抗噪双突破，堪称又快又聪明。
2
Sync to Async: Migrating FastAPI Endpoints to arq/Redis
从同步到异步，用arq/Redis优化FastAPI端点避免请求循环阻塞，提升LLM提取性能。
The problem: a PDF ingest that blocks for minutes My ingest endpoint was synchronous. Upload a file, wait for OCR, wait for the LLM extraction pipelin…
3
Show HN: DevQuest – live GitHub profile cards and a daily repository battle
实时生成GitHub个人资料卡片，还有每日仓库对战模式，让你的README动起来！
I built DevQuest, a tool for generating live GitHub profile cards for READMEs. I also added a Daily Battle mode where developers can submit a GitHub r…
4
阿里发布实时语音交互对话模型 Qwen-Audio-3.0-Realtime，号称懂倾听、更聪明
阿里发布实时语音交互模型，升级智商、共情与双工流畅度，号称更懂倾听更聪明。
IT之家 7 月 15 日消息，阿里巴巴实时语音交互对话模型 Qwen-Audio-3.0-Realtime 今天（15 日）正式发布，在智商、Agent 工具调用、共情对话和双工交互流畅度四条主线上同步升级，力求“又快又聪明”。 模型包括推理更强的 Plus 版本和速度更快的 Flash 版本，适…
5
STEC: Evidence Compression for Deep Search in Open-domain Multi-Hop QA
新方法STEC通过证据压缩提升开放式多跳问答的深度搜索效率与准确性
arXiv:2607.10795v1 Announce Type: new Abstract: In open-domain multi-hop question answering (QA), LLM-based search agents offer a promising approach t…
6
GPT-5.6-sol入榜DRACO：OpenSquilla集成方案仍在Brave组质量、成本双领先
DRACO评测揭晓：OpenSquilla多模型集成方案成本更低、质量更高，GPT-5.6-sol新入榜。
7
Show HN: Sql4json – Query JSON with SQL, no database
轻量级工具，无需数据库和配置，直接用 SQL 语法查询 JSON 数据，仅 500KB 大小。
Article URL: https://github.com/mnesimiyilmaz/sql4json Comments URL: https://news.ycombinator.com/item?id=48897784 Points: 1 # Comments: 0
8
I Built an AI That Turns GitHub Issues Into Pull Requests — No Local Setup Required
无需本地配置，AI自动将GitHub issues转化为PR，大幅提升代码维护效率，周转时间减少85%。
This is a submission for Weekend Challenge: Passion Edition What I Built resolvo is an agentic pipeline that takes a GitHub issue and a repo URL and h…
9
戴尔 Alienware 5K2K 165Hz 双模 QD-OLED 显示器 AW3926QW 发售，10999 元
39吋5K2K QD-OLED+165Hz双模，外星人旗舰显示器10999元开售
IT之家 7 月 13 日消息，Dell（戴尔）Alienware（外星人）现已在电商平台上架 AW3926QW 显示器。这一型号 采用 39" 1500R 的 RGB 条纹串联 OLED 面板 ， 支持 5120×2160 165Hz / 2560×1080 330Hz 双模切换 ，售价 1099…
10
QAgent: An LLM-based Multi-Agent System for Autonomous OpenQASM programming
LLM多智能体系统实现自主量子编程，QAgent能编写OpenQASM代码，探索AI与量子计算的融合。
arXiv:2508.20134v2 Announce Type: replace Abstract: Programming quantum circuits at the OpenQASM level is essential for achieving hardware-aware optim…
11
Testing LLMs Like Software: A Promptfoo Deep Dive for QA Engineers
深入探索Promptfoo工具，让QA工程师像测试传统软件一样高效测试LLM，附赠完整手册。
Want the full 46-page handbook? Promptfoo for QA: The Complete Engineer's Handbook (2026 Edition) by Himanshu Agarwal covers everything below in produ…
12
QQ 邮箱发布 AI 专用邮箱：Agent Mail，每人可抢注两个地址[内测]
QQ邮箱推出AI专用邮箱Agent Mail，支持命令行让AI工具自动收发邮件，每人限抢两个地址，内测中！
久违的 QQ 邮箱更新，发布了一款名为 Agent Mail 的 AI 邮箱服务，与个人 QQ 邮箱隔离，可以让你的 OpenClaw、Hermes、Claude Code、Codex 等工具收发邮件，实现与邮件有关的自动化。@Appinn 是不是可以让两个 AI Agent 通过邮箱聊天了？ AI
13
"The price table most AI cost tools multiply by has one automated test: jq empty
AI成本工具定价表仅依赖一个jq测试，历史价格被覆盖导致费用计算错误，点明数据管理隐患。
Most tools that tell you what your AI usage costs work the same way: count tokens locally, multiply by a shared price table. For most of the ecosystem…
14
Cognitive Defrag: A Local Sense Filter for Anti-Chaos Thinking (Zero-Data Edition)
用Python+aiogram打造本地文本去碎片化Telegram机器人，零数据泄露实现思维整理
The Manifesto: Against the "Cognitive Impasse" Let’s be honest. The tech world is currently drowning in poorly formulated requirements, chaotic task d…
15
Ask HN: What do you expect from a modern Go ORM?
Go 开发者探讨 ORM 新期望：如何平衡 SQL 控制权与查询简化？
I have been working on a Go ORM and I am trying to understand what experienced Go developers actually want from an ORM. The main problems I see: - los…
16
KQL Self-Study Day 3
KQL自学第三天实战分享，直面ADX界面输入的精准考验，查漏补缺的必修课。
Still working through the KC7 content. More of the same: following breadcrumbs through ADX. I think the biggest hurdle is fighting their interface - I…
17
How I Built a Bulk Video Processing Pipeline with Next.js, BullMQ, and FFmpeg
用Next.js+BullMQ+FFmpeg搭建并行视频处理管道，50个视频同时去静音、加字幕、裁剪，服务器不崩的实战方案
I've been building BatchEdits — a tool that lets content creators upload 50 videos at once and get them back edited simultaneously with silence remove…
18
Show HN: Cyrinx (36kbps Acoustic Transport)
通过声学信道实现36kbps数据传输，利用OFDM和16-QAM调制，自动消除时钟漂移与相位噪声，适合离线或低功耗场景通讯
Working with Fable 5, a MacBook, and a Pixel phone, I built an acoustic transport library that is several orders of magnitude faster than existing SoT…
19
Show HN: Handsum: An LQIP Image File Format
全新开源图像格式Handsum专为低质量占位图优化，大幅提升Web图片加载体验
Article URL: https://nigeltao.github.io/blog/2026/handsum.html Comments URL: https://news.ycombinator.com/item?id=48837153 Points: 2 # Comments: 0
20
AI 大模型周榜：claude-fable-5 蝉联榜首，国产模型稳定中上游
AI大模型周榜出炉，Claude蝉联榜首，国产模型稳定在中上游，美国厂商包揽前十。
IT之家 7 月 9 日消息，Arena（ arena.ai ）平台公布了最新一期大模型对战周榜，本期统计覆盖 2026 年 6 月 29 日 ~7 月 5 日。 本周榜首位置依旧由 🇺🇸 Anthropic 的 claude-fable-5 牢牢占据 ，分数微涨 1 分，美国厂商在头部占据绝对优势…