AI BriefingXing4.0-29B-A4B-GGUF: 262K Context Support, GGUF Quantization of the 29B MoE Model
Venastine-Research/Xing4.0-29B-A4B-GGUF
About the project
China Telecom's Xing4.0 29B MoE model has been released in GGUF format. Out of 29B total parameters, only 4B are active, enabling efficient inference in local environments.
Various quantization versions are provided, ranging from IQ2 to Q8 and f16. It supports a long context window of 262,144 tokens, with tool calling and reasoning modes included in the base template.
Licensed under Apache 2.0, commercial use is permitted. It can be run directly in GGUF-compatible environments such as llama.cpp, allowing users to experience the performance of large-scale models without high-spec GPUs.
Venastine-Research/Xing4.0-29B-A4B-GGUF
The original page has no description.
text-generation
This introduction was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report errors, attribution issues, or removal requests via Contact.