Qwen3.5-2B 端侧 LiteRT-LM 量化版发布
原标题：litert-community/Qwen3.5-2B
AI 摘要
litert-community 将阿里 Qwen3.5-2B 转换为 LiteRT-LM 格式，发布纯文本与视觉两个 int8 量化版本，用于端侧推理。在 Galaxy S26 上，GPU 解码速度达 17.5 tok/s，CPU 达 19.0 tok/s；在树莓派 5 上，CPU 解码速度为 4.2 tok/s。作者承认此前量化配方存在缺陷，导致 CPU 后端在多步算术上出现循环回复，并已在 2026-10-11 重新量化修复，同时替换了分词器以修正组合字符的分词差异。
正文节选
--- license: apache-2.0 base_model: Qwen/Qwen3.5-2B tags: - litert - litert-lm - litertlm - on-device - edge - hybrid - gated-deltanet - qwen3_5 - vlm - image-text-to-text pipeline_tag: text-generation library_name: litert-lm base_model_relation: quantized --- LiteRT is Google's on-device runtime, the new name for TensorFlow Lite (Android: `com.google.ai.edge.litert:litert`), and `litert-torch`, the renamed `ai-edge-torch`, is its PyTorch converter: a PyTorch model converted