XingChen-AGI/Xing4.0-29B-A4B-GGUF official quant repo

Official GGUF companion to XingChen-AGI/Xing4.0-29B-A4B, verified against the HF API and the raw README on 2026-10-17. HF API createdAt 2026-09-16T13:23:49Z, lastModified 2026-10-16T04:13:39Z. API gguf block: architecture xing4_0, total 31,215,031,088, context_length 262144, totalFileSize 15,446,399,712 (one file, not the sum of the repo). Siblings: xing4_0-29b-iq3_xxs.gguf, iq3_m.gguf, iq4_xs.gguf, iq4_nl.gguf, IQ4_NL.gguf, q5_0.gguf, q6_k.gguf, q8_0.gguf. The README says the weights use IQ4_NL mixed-precision quantization, that the quantized GGUF is approximately 18 GB, and that it can run on a single consumer-grade GPU; it points at the llama.cpp tutorial in the Xing4.0 GitHub repo. It does not publish per-quant resident-set measurements, so none are recorded here. Repo usedStorage is 188,573,421,568 bytes. Apache-2.0.