OpenAI官方指南，详解GPT-5.6如何驱动创业公司打造更省钱、更高效的AI Agent，掌握智能模型选择与Responses API新特性。
Learn how startups use GPT-5.6 to build faster, more cost-efficient AI agents with smarter model selection and new Responses API capabilities.
医疗影像中选预训练模型总靠试错？本文系统检验了迁移性估计指标在医学场景下的鲁棒性，帮你少走弯路。
arXiv:2608.09999v1 Announce Type: cross Abstract: In transfer learning, the choice of source model largely influences the performance on a target data…
54个LLM工作流如何选对模型？从Bedrock到OpenRouter的实战调优思路。
I have a Django app with 54 LLM-backed workflows. Up until recently I've exclusively used Anthropic models via AWS Bedrock but just set up OpenRouter …
面对OpenRouter上琳琅满目的AI模型，如何做出明智选择？这里有实战经验与决策思路分享。
I am experimenting with models other than Claude or OpenAI, using OpenWebUI connected to openrouter.ai and am overwhelmed by the choices. The descript…
Ethan Mollick 的 AI 选择指南：从聊天模型进化到 agent，帮你快速选对工具。
An opinionated guide to which AI to use to do stuff It's interesting watching the evolution of Ethan Mollick's guide over time. A year ago it was stil…
本地开源的大型语言模型费用优化工具，自动识别可用廉价模型替代的高成本调用，让AI支出更省心。
Hi HN. I built Frugon to answer one question on my own machine: which of my LLM calls could I run through a cheaper model and stretch my dollar withou…
别凭记忆选模型，用实时数据说话：215个模型、7个免费候选，证据驱动的选型指南。
Most teams do not need the perfect AI model on day one. They need a first model they can explain. The mistake is starting from brand memory: choose a …
开发者必看！一文搞懂Claude Opus和Haiku的核心区别，助你选对模型降本增效。
Anthropic's Claude family has grown into a diverse lineup of models optimized for different workloads. Rather than having a single "best" model, each …
弱监督ASR数据质量难保证？新方法用过滤与选择机制提升大规模训练效果，附实验验证。
arXiv:2606.28728v1 Announce Type: cross Abstract: Leveraging large-scale weakly supervised datasets is crucial to train robust end-to-end automatic sp…
追踪百万次LLM调用后发现六成用错模型，选对模型能省大钱！
Article URL: https://tokonomics.ca/blog/we-tracked-1m-llm-api-calls-most-were-wasting-money Comments URL: https://news.ycombinator.com/item?id=4870866…
针对不同文章类型自动切换最优LLM模型，并集成80+正则清洗反AI痕迹，提升内容真实感。
I've been quietly running a small network of niche content sites for a few weeks. Two are live on real domains: afixu.com - home improvement and DIY t…
开发者晒出本地跑大模型的硬件与模型配置，实战经验直接参考。
Comments URL: https://news.ycombinator.com/item?id=48698057 Points: 3 # Comments: 1
GitHub Copilot免费版与学生版取消手动选模型，改为系统分配，配合按量计费新规引发开发者关注。
IT之家 6 月 25 日消息，GitHub 于当地时间 6 月 24 日宣布，Copilot Free 与 Student 计划现已将 Copilot 自动模型选择作为默认且唯一的模型选择方式 。 GitHub 宣称，自动模式会为各项任务动态挑选最佳模型，免去了用户手动选择的繁琐。在受限于各套餐规…
将LLM提示智能路由到最便宜且能胜任的模型，大幅降低AI API调用成本。
Article URL: https://github.com/Reactance0083/pydantic-ai-multi-llm-cost-optimizer Comments URL: https://news.ycombinator.com/item?id=48644688 Points:…
手把手教你将Reasonix 1.x与DeepSeek V4集成，实战解析ACP模型选择器配置要点。
Integrating Reasonix 1.x with DeepSeek V4: ACP Model Selector Integration in Practice This article discusses how to switch Reasonix 1.x, a local ACP C…
工厂自动选模型，降本提效高达25%，让LLM路由像呼吸一样简单
Article URL: https://factory.ai/product/router Comments URL: https://news.ycombinator.com/item?id=48575781 Points: 1 # Comments: 0
一个开源智能路由，自动为LLM请求选择最优模型，省去手动调用的麻烦。
Article URL: https://github.com/iliatankelevich/maslul Comments URL: https://news.ycombinator.com/item?id=48581123 Points: 2 # Comments: 0
自动化选择视觉语言模型预训练方案，提升多模态任务效率
arXiv:2502.00241v2 Announce Type: replace-cross Abstract: Incorporating multiple modalities into large language models (LLMs) is a powerful way to enh…
24GB显存下寻找超越Qwopus3.6的8-bit LLM模型，生产可用性优先。
What’s the best model right now that outperforms Qwopus3.6-27B-v2-MTP-GGUF 8-bit on a 24 GB VRAM GPU? Looking for real reviews. I found 4 bit not usab…
微软CEO纳德拉坦言：AI不是万能钥匙，盲目追求最强模型反而可能造成资源浪费，微软内部已开始反思与管控。
IT之家 6 月 11 日消息，萨提亚 · 纳德拉想对微软一众沉迷人工智能的员工说：并非所有问题都要动用性能最强的 AI 模型。 在《纽约时报》旗下播客《Hard Fork》的现场录制活动中，有人向这位微软首席执行官提问，公司内部如今盛行多大程度的算力堆砌（tokenmaxxing）行为。 主持人凯…