开源工具
x.com ·
@Thom_Wolf
Aikido 发布开源权重安全模型 Altar-1，基于 GLM 5.3 剪枝量化
Thomas Wolf 转述 Aikido 推出首个开源权重安全模型 Altar-1，称其为开源 SOTA 模型 GLM 5.3 的剪枝量化版本，轻量到可部署在单台 4×H200 节点上，并认为开源防御性安全模型正接近前沿水平。该评价为主观判断，未附基准数据。
AI推荐理由 Hugging Face 联合创始人 Thomas Wolf 转述 Aikido 发布开源权重安全模型 Altar-1，称其为 GLM 5.3 的剪枝量化版本，可装入单台 4×H200 节点，指向开源防御性安全模型接近前沿的趋势。信息为转述加简短评论，缺少基准数据与部署细节，'接近前沿'属主观判断。
66
浏览 1,940
点赞 12
评论 3
转发 3
正文
nice to start seeing more open-weight security model with defensive capabilities which are close to the frontier
altar-1 from Aikido is a pruned and quantized version of the open-SOTA GLM 5.3 which is lightweight enough to fit on one 4-H200s node
Aikido@AikidoSecurity
Introducing Altar-1, our first open-weight security model.
Frontier-grade defensive AI, built to deploy. Own your own security.
[Your browser does not support the video tag.](https://video.twimg.com/amplify_video/2102031856137019392/vid/avc1/1280x1280/IVLP1nelqyiEFH2t.mp4?tag=29)
[🔗 View on Twitter](https://x.com/AikidoSecurity/status/2102035136678400056/video/1)
[🔗 View Quoted Tweet](https://x.com/AikidoSecurity/status/2102035136678400056)