Unsloth 量化 Thinking Machines Inkling，280GB 可运行
You can now run Thinking Machines Inkling!
Unsloth 宣布可以运行 Thinking Machines 的 Inkling，并将其量化为 Dynamic 1-bit，体积减少 86%，保留了 top-1% 准确率的 74.2%，可在 280GB 上运行。Inkling 是 975B 开源模型，支持图像、音频和 1M 上下文。Unsloth 同时提供了使用指南和 GGUF 文件。
推荐理由
精选AI 评分 63
Unsloth 给出 Inkling 的量化版本与体积、精度保留数据，读者可据此判断本地部署门槛。
主题
查看原帖Today, we are introducing Inkling. Inkling reasons efficiently across text, image, and audio modalities. We are making the full weights available. https://t.co/Ghebq5mG30 Available today for fine-tuning on Tinker. Play with it in the Inkling Playground. 🧵
正文
当前显示原文，中文全文译文尚未就绪。
You can now run Thinking Machines Inkling!
Inkling is a 975B open model with image, audio and 1M context support.
We quantized Inkling to Dynamic 1-bit (-86% size) and retained 74.2% of top-1% accuracy. Run on 280GB.