Gemma 4 12B 推出 LiteRT-LM 格式，支持多模态与多令牌预测
原标题：litert-community/gemma-4-12B-it-litert-lm
AI 摘要
litert-community 发布了 Gemma 4 12B 模型的 LiteRT-LM 格式版本，支持文本、视觉和音频模态以及多令牌预测，可在 macOS、Linux、Windows 和 Web 上部署。该模型基于 Google 的 Gemma 4 12B，适合桌面端离线使用，需 LiteRT-LM v0.17 或更高版本。模型文件大小约 6.9GB，在 NVIDIA 4090 上解码速度达 69 tokens/s，在 MacBook M4 Pro 上为 29 tokens/s。
正文节选
--- license: apache-2.0 base_model: - google/gemma-4-12B-it tags: - litert-lm --- # litert-community/gemma-4-12B-it-litert-lm Main Model Card: [google/gemma-4-12B-it](https://huggingface.co/google/gemma-4-12B-it) This model card provides the Gemma 4 12B model in LiteRT-LM format ready for deployment on macOS, Linux and Windows, as well as web in a more limited capacity. The current [`gemma-4-12B-it.litertlm`](https://huggingface.co/litert-community/gemma-4-12B-it-litert-lm/blob/main/gemma-