模型中心DoubleAI代码执行模型性能优化Hacker News模型发布

Maxwell: A new kind of model that sits between code and hardware

来源:Hacker News 首页收录:2026/10/19

✦ AI 解读

DoubleAI 在 Hacker News 发布 Maxwell,一种介于代码与硬件之间的新型前沿模型,称为“代码执行模型”。它无需人工介入,能在不修改代码的前提下让程序运行更快。Maxwell 通过追踪、性能分析和数据流从内部理解程序,并动态适应硬件、运行环境及输入变化。团队认为,当前针对不同用例反复优化代码(如 vLLM 等 LLM 推理引擎)的方式不必要、次优且难以扩展,Maxwell 旨在改变这一现状。

Hi HN, I'm Gal from doubleAI. We've built Maxwell, a new kind of frontier model we call a Code Execution Model.It sits between your code and the hardware. No humans in the loop! Maxwell makes the code run faster as it runs, without ever changing a single line.Maxwell sees programs from the inside: traces, profiling information, and the data flowing through them. It adapts dynamically to the environment, including the hardware, what else is running, and what the inputs look like.Code execution models change the way we approach coding. Today, Claude/Codex (and humans) keep re-optimising code for every different use-case hardware and data flow (think of LLM inference engines, like vLLM!). Our bet is that this is unnecessary, suboptimal, and that it does not scale. Instead, we think the right answer is to make code a living, thinking, thing.Early access is open at heymaxwell.ai. Happy to answer questions here. Comments URL: https://news.ycombinator.com/item?id=50152466 Points: 16 # Comments: 7
查看原文 ↗

相关推荐

Kimi K3.1 Pro and K3.1 Pro Highspeed
✦ AI 摘要月之暗面(Moonshot AI)在 Kimi Code 文档中上线 Kimi K3.1 Pro 与 K3.1 Pro Highspeed 两个模型版本,前者主打高质量代码生成,后者面向低延迟高速推理场景。该消息在 Hacker News 引发 169 点、79 条评论讨论,从业者关注其代码能力、定价与高速版的实际吞吐表现。
Kimi K3.1 Pro and K3.1 Pro Highspeed
✦ AI 摘要Kimi 发布 K3.1 Pro 与 K3.1 Pro Highspeed 两款模型,官方文档页面已上线,并在 Hacker News 引发讨论(86 分、21 条评论)。Highspeed 版本推测主打低延迟推理,适合对响应速度敏感的编码等场景。目前公开信息有限,具体参数、基准评测与定价尚未披露,建议关注官方后续更新。
Gemma-4-26B-A4B with 39-69% less thinking at same accuracy
✦ AI 摘要Hacker News 上分享的一篇文章介绍 Gemma-4-26B-A4B 模型,声称在保持相同准确率的前提下,将思考(推理)过程减少了 39% 至 69%。该内容来自个人博客 bottlecapai.com,讨论热度较低(3 分、0 评论),但涉及大模型推理效率优化这一从业者关注的方向,具备一定参考价值。