Ingot子网获胜量化模型Ingot-VI-27B-Q2公开镜像
原标题：dendriteholdings/Ingot-VI-27B-Q2
AI 摘要
Hugging Face 上发布了 Ingot-VI-27B-Q2，这是 Bittensor 子网 Ingot 上获胜模型的公开镜像。该子网让矿工对固定的教师模型 Qwen/Qwen3.8-27B 进行量化，验证者选出与教师模型最接近的副本。该模型由矿工 uid 121 提交，在获胜评估中 KL 为 0.10883、隐藏状态距离为 0.3356，优于此前冠军，并比 genesis 版本分别接近 16.0% 和 26.1%。模型为 GGUF 格式，需使用 PrismML 的 llama.cpp fork 运行。
正文节选
--- base_model: Qwen/Qwen3.8-27B base_model_relation: quantized library_name: llama.cpp pipeline_tag: text-generation tags: - gguf - ingot - bittensor --- # Ingot-VI-27B-Q2 A public mirror of a model that won on Ingot, a Bittensor subnet where miners quantize a fixed teacher model and the validator crowns the copy that stays closest to it. - Submitted by miner hotkey `5Dk5NHaBze5mex99s276kAPb66HouXZ7Y7MMgb7Azu1pHUnc` (uid 121) - Crowned from submission `67693dc4-e2d7-4b00-af1a-2b42bf767fb4`