Meta 发布 Muse Spark 1.3，官方称模型未经基准刷分优化
Here are the benchmarks. As always, these are numbers... and now it's time to test it on real use c...
精选理由
Meta 放出了 Muse Spark 1.3 的跑分，重点是官方说没刷榜，实际用起来怎么样可以自己试试。
Meta 发布 Muse Spark 1.3 模型并公布了基准测试数据。发布者 Paul 在 X 上指出，Muse Spark 1.3 没有为基准测试做针对性优化（benchmaxx）。详细数据可在 Meta 的研究博客 research.meta.ai 查看。作者建议把模型放到实际使用场景中验证表现。
原文 · Paul Couvert
Here are the benchmarks. As always, these are numbers... and now it's time to test it on real use c...
Here are the benchmarks. As always, these are numbers... and now it's time to test it on real use cases. But Muse Spark 1.3 wasn't benchmaxxed so let's see. Full details in their blog post: research.meta.ai/blog/muse-wond… 💬 0 🔄 0 ❤️ 2 👀 562 ⚡
