Gemma 4 26B A4B 推出 LiteRT-LM 量化版
原标题：litert-community/gemma-4-26B-A4B-it-litert-lm
AI 摘要
Hugging Face 上发布了 litert-community/gemma-4-26B-A4B-it-litert-lm，这是 Google Gemma 4 26B（A4B）混合专家模型的 LiteRT-LM 量化版本，采用 QAT 的 blockwise int4 权重与 float 激活，面向 Web 和桌面端本地部署。该版本目前仅支持文本，音频、图像与多 token 预测将在后续更新中加入。模型卡给出了 Linux、macOS、Windows 和 Web 上的性能基准，并说明桌面端需 LiteRT-LM v0.17.1 或更高版本。
正文节选
--- license: apache-2.0 base_model: - google/gemma-4-26B-A4B-it base_model_relation: quantized tags: - litert-lm --- # litert-community/gemma-4-26B-A4B-it-litert-lm Main Model Card: [google/gemma-4-26B-A4B-it](https://huggingface.co/google/gemma-4-26B-A4B-it) This model card provides the Gemma 4 26B (A4B) mixture-of-experts model in LiteRT-LM format that is ready for deployment on web and desktop. Please check back here regularly for updates on wider platform support and further functionali