GGUF conversions of the tiny Kimi-K3 0.40B development checkpoints. — Kimi-K3 architecture validation; — llama.cpp conversion and inference testing; — MXFP4 MoE layout testing; — out-of-core expert-loading research. They are not intended as useful language models and should not be treated as quality proxies for the full Kimi-K3 model. — 168 routed expert projection tensors remain in MXFP4; — they are repacked into 21 GGUF expert groups; — 35 resident MoE tensors are dequantized to F16; — unknown packed tensors are rejected by the converter. — repository: murillo128/llama.cpp — branch: k3/out-of-core — converter commit: 84245db4c790af22135f34992689edcc11877003 — exact source revisions; — generated file sizes; — SHA-256 checksums; — conversion counts; — Python package versions. Both F16 and hybrid MXFP4 fixtures were loaded and executed successfully using the CPU backend. The observed fixture continuation was: This is only a loading and execution smoke test, not a correctness or quality benchmark. These artifacts support the k3-out-of-core project, which explores tiered resident, RAM, and NVMe storage for routed MoE experts in llama.cpp.
Модальности:
Генерация текста
Задача: Генерация текста
Автор: murillo2000
Теги: llama.cpp, gguf, kimi-k3, mxfp4, moe, test-model, endpoints_compatible
Лайков: 4 | Загрузок: 2,424
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
