Event ·Multi-source

Bindu Reddy：权重空间编程或让小模型性能逼近大模型

Bindu Reddy 在 X 上发文称，其团队用 Astra 直接修改小模型的权重来提升推理性能，这种方法被称为权重空间编程。她认为大模型未来可以对小模型做精细化的权重更新，缩小两者在常见任务上的差距。她同时预测这一方向会把 token 价格再压低 10 倍，DeepSeek 这类模型会更受欢迎。

Current read

Arcee 前高管 Bindu Reddy 分享用 Astra 直接改小模型权重提升推理的思路，还预测 token 价格会再降 10 倍，观点挺有意思。

11 sources34° AI heatLast updated 2026/11/3 06:12:07

Evidence

11 sources

Primary sourceBindu Reddy

2026/11/3 06:12:07View source

Related source 1kimmonismus

2026/11/1 10:51:09View source

Related source 2The Information

2026/11/2 15:00:25View source

Related source 3Latent.Space

2026/11/3 05:40:15View source

Related source 4Jerry Liu

2026/11/1 17:40:15View source

Related source 5Thomas Wolf

2026/11/2 20:15:21View source

Related source 6rohanpaul_ai

2026/11/2 20:17:28View source

Related source 7The Rundown AI

2026/11/2 20:17:39View source

Related source 8orange.ai

2026/11/2 23:30:19View source

Related source 9宝玉

2026/11/3 00:06:53View source

Related source 10shao__meng

2026/11/3 00:45:13View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story