Kimi-K3-Audit-451E is Kimi K3 specialised for security auditing. It keeps 451 of the 896 routed experts in each layer, which brings the model onto two 512 GB Apple silicon machines at half the size of the release. On kernel-audit workloads it measures at parity with the full model. Expert selection was calibrated on kernel source and audit traces, targeting domain focus rather than compression. For a single 512 GB machine, use the smaller Kimi-K3-Audit-240E build instead. Security auditing and analysis of source code, with tool use. General conversation, translation and non-technical writing are out of scope, and a general-purpose model should be used for them. Across eight code corpora in five languages, pruning changes perplexity by at most 1.10×, against 1.38× for the 240E build. Measured against the original MXFP4 release (896E) on identical tokens. KLD is KL(896E ‖ 451E) over the full output distribution. Each of 24 kernel-audit sessions contributes one 1025-token window, rendered through the model’s own chat template so the structure matches deployment. Overall perplexity on audit traces is slightly below the release. Tool-call spans diverge least of any span measured. 81%…
Модальности:
Генерация текста
Задача: Генерация текста
Автор: catalystsec
Теги: mlx, kimi_k3, kimi-k3, moe, expert-pruning, code-audit, custom_code, en
Лайков: 3 | Загрузок: 681
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
