手机端首次跑通27B大模型，Bonsai利用量化技术让AI真正落地移动设备。
IT之家 7 月 15 日消息，@PrismML 官方账号今天（7 月 15 日）发布博文，宣布推出 Bonsai 27B 模型，基于 Qwen 3.6 27B 模型微调， 在保留 90% 智能水平的情况下，可以在 12GB 内存的 iPhone 上原生运行。 Qwen 3.6 27B 模型进一步提…
一款基于开源构建的股票分析工具OpenClaw，助你快速洞察市场动态。
I'm the solo dev/builder of FN2, which uses Open Source AI models to deliver autonomous agents at scale, similar to Hermes and OpenClaw but tailored f…
IT之家 7 月 15 日消息，昨日（7 月 14 日）在接受 CNBC 采访时，PrismML 首席执行官巴博克 · 哈西（Babak Hassibi）表示， 其公司确实和苹果公司接洽，并透露苹果公司正在评估其相关 AI 量化技术。 在采访中，哈西表示：“他们 [指苹果公司] 目前正在认真评估我们…
针对长思维链大语言模型的KV缓存量化新突破，渐进混合精度方案，已被ICLR 2026接收，代码已开源。
arXiv:2505.18610v2 Announce Type: replace Abstract: Recently, significant progress has been made in developing reasoning-capable Large Language Models…
量化后的LLM看似准确率不变，但推理过程已悄然“空心化”，本文用六类故障图谱揭示这一隐患。
arXiv:2607.09999v1 Announce Type: cross Abstract: We show that post-training quantization can silently alter how large language models reason even whe…
用LLM驱动代码进化，从高维金融数据中挖掘隐藏的预测信号，突破传统因子发现瓶颈。
arXiv:2511.18850v4 Announce Type: replace Abstract: Discovering effective predictive signals, or "alphas," from financial data with high dimensionalit…
新研究揭示量化大语言模型的可靠性缩放定律，为部署经济高效且可信赖的AI模型提供理论依据。
arXiv:2607.10855v1 Announce Type: new Abstract: Quantization is a powerful strategy to build capable and resource-efficient large language models (LLM…
IT之家 7 月 14 日消息，今日腾讯混元发文表示，Hy3 上线并开源后，社区一直在呼吁量化版本，希望单卡跑参数量达 295B 的 Hy3。团队针对开源社区需求，把 Hy3 的权重量化到 1bit 与 4bit 并打包成 GGUF，配合 llama.cpp 生态，让原本只能跑在多卡集群上的模型，最…
无需隐变量，三值量化LLM训练新方法，开源项目BitBop让你以极低成本体验高效训练。
Article URL: https://github.com/ValerioDolci/bitbop Comments URL: https://news.ycombinator.com/item?id=48892013 Points: 1 # Comments: 1
从Jetson Nano到模型变体，揭示边缘LLM基准测试中易忽视的变量，避免自欺欺人。
A Jetson Nano and Ollama benchmark review appeared among DEV's top articles on 2026-07-12 . Edge inference experiments are valuable because they make …
基于谷歌开源LLM打造的专用向量化AI助手，论文详解模型设计与应用场景。
arXiv:2607.07846v1 Announce Type: new Abstract: VectorizationLLM is a specialized Large Language Model based on Google open-weight LLMs. The model is …
设备端AI推理新突破：自动云回退、4-bit无损量化、PyTorch模型一键转换。
Hi HN, Roman and Henry here from Cactus ( https://github.com/cactus-compute/cactus ). We just shipped the biggest upgrade to our on-device inference p…
开发者亲述如何在RTX 3050消费级显卡上打造完全离线、隐私优先的AI编排器，实现亚100ms推理速度，兼顾性能与隐私。
Hi, I'm Raj Patil (known online as Dream / lostxmusafir ), an AI Engineer and Full-Stack Developer. In this article, I want to break down how I develo…
用Kronecker分解海森矩阵，高效提升LLM量化精度，突破后训练量化瓶颈。
arXiv:2607.07964v1 Announce Type: new Abstract: Post-training quantization (PTQ) is a widely adopted technique for compressing large language models (…
在慢速电脑上运行744B参数的GLM-5.2模型，实测不同精度下多token预测头的性能表现。
A few days ago I found myself trying out GLM 5.2 and was really positively impressed. The capabilities and security I was getting from this LLM are si…
AI编码代理也能炒股票了，这款CLI直接接入实时金融数据，让大模型帮你做股票、期权、SEC文件深度研究。
Hi, my name is Kam, and today my cofounder Josh and I are shipping Finterm.ai, a CLI that gives coding agents direct access to financial data: stock p…
新型LLM量化方法FPTQuant，通过功能保持变换解决离群值问题，高效节能不损性能。
arXiv:2506.04985v2 Announce Type: replace Abstract: Large language models (LLMs) require substantial compute, and thus energy, at inference time. Whil…
年营收仅3000万，估值却高达140亿美元，资本狂赌物理世界基础模型成为机器人时代的底层操作系统。
前沿大模型突破容器沙箱防线的能力首次被量化评估，揭示AI安全新风险
arXiv:2603.02277v2 Announce Type: replace-cross Abstract: Large language models (LLMs) increasingly act as autonomous agents, using tools to execute c…
首次实现全栈FP4量化预训练，突破LLM训练速度和显存瓶颈
arXiv:2607.04422v1 Announce Type: cross Abstract: Recent NVFP4 pretraining methods mainly target transformer linear layers, leaving optimizer states, …