1
Show HN: InferAll – One API for OpenAI, Anthropic, Google, Nvidia Nim
统一API接入多家大模型，更可免费使用118+开源模型，仅需$5激活。
Article URL: https://inferall.ai/?src=hn Comments URL: https://news.ycombinator.com/item?id=48920849 Points: 1 # Comments: 0
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
GitHub is starting to lockdown some of the previously public data
GitHub突然隐藏了星标追踪数据，开发者毫无预兆地被“静默”限制API访问，平台政策转向引发不安。
I woke up to an alert about a database anomaly: one of the tables that we use to track stargazers was suddenly empty. Turns out GitHub quietly made th…
4
Saving Money on AI APIs? Start With These 30 Models
30个低成本AI模型推荐，实验想法只需几毛钱，适合预算有限的开发者快速测试。
Saving Money on AI APIs? Start With These 30 Models I'll be honest with you. When I finished my coding bootcamp, I thought I had this whole AI thing f…
5
I Built AICostPass Because I Was Tired of Guessing My AI API Costs
精准追踪AI API成本，告别盲目猜测，这款新工具让开支清晰可见。
While building with OpenAI, Anthropic, and other AI providers, I realized something surprising. I monitored my servers, databases, and application per…
6
React Native + Expo JWT Auth with FastAPI — A Complete Working Example
从零搭建Expo移动端与FastAPI后端的JWT认证，并附完整可运行的代码示例，手把手教你处理access/refresh token与OAuth2集成。
Most FastAPI JWT tutorials stop at "here's how you issue a token" and never cover the parts that actually break in production: where to store the toke…
7
Show HN: Low-latency local LLM runner via OpenJDK Panama FFM (Java 22)
基于OpenJDK Panama FFM的Java 22低延迟本地LLM运行器，统一硬件编排与零分配ABI，线程安全C API实现零拷贝编译。
I wanted to run AI from inside the JVM. I started out with the standard REST sidecar, ripped that out to use Project Panama (Foreign Function & Memory…
8
Bootcamp Grad Explores Open-Source AI APIs: What I Learned
从训练营毕业生的视角，揭秘开源AI API的成本与体验，告诉你为何自建模型可能比API更贵。
Here's the thing: bootcamp Grad Explores Open-Source AI APIs: What I Learned I graduated from a coding bootcamp about six months ago, and honestly, I …
9
Pritset: Generate PDFs using DOCX files as templates via API
用DOCX模板作身份证，API快速生成PDF，支持直接请求与webhook双模式。
I've designed and create Pritset to make it easy to generate tons of of PDFs based on your DOCX templates. There are two flows: direct and webhook. Fi…
10
Agent要数量也要脑子！浪潮信息一边单柜养4万Agent，一边让大模型组队答题
单柜养4万Agent，还让大模型组队答题，浪潮信息这次把AI智能体规模与智商双双拉满。
同时发布CPU原生液冷整机柜、多模融合超节点
11
DeepSeek cut prices 75%. The 100x problem remains
DeepSeek降价75%仍难解困局，OpenAI以2亿美元API额度笼络YC初创，AI定价战背后的商业逻辑值得深究。
DeepSeek's recent decision to drastically cut pricing on its V4-Pro model by 75% should have been unequivocally good news for enterprise AI vendo…
12
苹果披露正为欧盟地区 iOS 系统开发全新 API：允许第三方产品跨设备自动同步配对、首批产品支持 Meta Quest 头显
为欧盟iOS开发的新API让第三方设备像AirPods般跨设备自动同步配对，首批支持Meta Quest头显
IT之家 7 月 13 日消息，据外媒 MacRumors 报道，苹果在欧盟《数字市场法》互操作性页面中披露公司正为 iOS 设备开发一项全新的配件配对 API，让第三方设备也能像 AirPods 或 Apple Watch 一样能够在用户的所有苹果设备之间实现自动同步配对， 首批产品将包含 Met…
13
Making an email API agent-native: llms.txt, a /api/me endpoint, and a zero-DNS sandbox sender
让邮件API原生适配AI智能体：统一端点、llms.txt与零DNS沙箱发送器。
Most APIs are "agent-compatible": an agent can call them. Far fewer are agent-native : an agent can accomplish the whole job on its own, with no human…
14
frp v0.70：开源内网穿透工具教程，完善 API、规定兼容性策略
frp v0.70 大版本更新：API v2 迁移、WebSocket 兼容性增强，内网穿透神器再进化
frp v0.70 更新，完成了 frps 的 Dashboard API v2 迁移、提升了 WebSocket/WSS 兼容性、修复了 Proxy Protocol 问题。v0.69 还规定了新的兼容性策略。@Appinn frp v0.70 这个版本主要继续推进 frps Dashboard
15
7 things I learned trying to stop LLM API bills from silently exploding
通过标签追踪与重试策略调优等7个实战技巧，阻止你的LLM API账单在后台悄悄燃烧。
My first real LLM bill surprise wasn't dramatic. No infinite loop, no viral spike. A retry policy I'd written months earlier met a flaky endpoint, and…
16
1662 万美元！开发者收到 Anthropic 巨额 AI 账单，Claude 控制台显示 0 美元
实时监控AI API使用量，智能预警防止天价账单，支持Claude等主流模型费用追踪与异常检测
IT之家 7 月 11 日消息，科技媒体 Wccftech 昨日（7 月 10 日）发布博文，报道称 一位韩国开发者收到来自 Anthropic 的 160 万美元、1662 万美元两笔扣款账单 ，但在 Claude API 控制台中显示为 0 美元。 博文指出该开发者于 7 月 8 日共收到 2 …
17
Building a Fully Automated Facebook Post Scheduler using Node.js and GitHub Actions
教你用Node.js和GitHub Actions实现Facebook自动化发帖，永久令牌配置+脚本全流程拆解。
How I Built a Zero-Cost Facebook Auto-Poster Using Node.js and GitHub Actions Automating social media management can save hours of manual work. In thi…
18
Show HN: Inferock-bench – per-call billing receipts for OpenAI and Anthropic
开源工具，为OpenAI和Anthropic API提供逐次调用的计费收据，轻松管理账单。
Article URL: https://github.com/inferock/inferock-bench Comments URL: https://news.ycombinator.com/item?id=48868354 Points: 1 # Comments: 0
19
Do not migrate an AI API by changing only the base URL
仅改base URL迁移AI API？小心陷阱！深度解析兼容性迁移的真正要点
Changing the base URL is the easy part of an OpenAI-compatible migration. The real migration starts when the first request has to be explained. If a t…
20
Meta 正式向开发者开放 Muse Spark AI 模型 API，加入 AI 商业化竞争
Meta 向开发者开放 Muse Spark AI 模型 API，专注编程与智能体任务，提供 20 美元免费体验额度，按量付费。
IT之家 7 月 9 日消息，Meta Platforms 今日正式向开发者开 Muse Spark AI 模型 API，同时发布了升级版 Muse Spark 1.1。此举意味着 Meta 正式加入 AI 模型商业化竞争，与 Anthropic、OpenAI 等公司一样，开始通过模型调用收费。 M…