---
title: Qwen3.5-2B-LiteRT
canonical_url: "https://www.modelscope.cn/models/litert-community/Qwen3.5-2B-LiteRT"
md_url: "https://www.modelscope.cn/models/litert-community/Qwen3.5-2B-LiteRT.md"
repository: litert-community/Qwen3.5-2B-LiteRT
last_updated: 2026-04-05
license: apache-2.0
pipeline_tag: image-text-to-text
tasks:
  - image-text-to-text
library_name:
  - pytorch
frameworks:
  - pytorch
downloads: 35
stars: 1
tags:
  - litert
  - tflite
  - on-device
  - qwen
  - qwen3.5
---

# Qwen3.5-2B-LiteRT

> Qwen3.5-2B-LiteRT - litert-community 在 ModelScope 开源的模型。qwen3.5-2b local multimodal conversion

litert-community/Qwen3.5-2B-LiteRT 是 ModelScope 魔搭社区上的image-text-to-text模型，采用 apache-2.0 许可。

- **Repository**: litert-community/Qwen3.5-2B-LiteRT
- **License**: apache-2.0
- **Tasks**: image-text-to-text
- **Tags**: litert, tflite, on-device, qwen, qwen3.5
- **Downloads**: 35
- **Stars**: 1
- **Last updated**: 2026-04-05

Source: https://www.modelscope.cn/models/litert-community/Qwen3.5-2B-LiteRT

---

# qwen3.5-2b local multimodal conversion

This repository contains a locally converted Qwen3.5 multimodal LiteRT package.

Locally converted Qwen3.5 2B multimodal LiteRT package with split vision/text components.

- target: `qwen3.5-2b`
- artifact: `model_multimodal.litertlm`
- size: `2656.62 MB`
- generated_at: `2026-04-04T12:43:19.072877+00:00`

## Package Contents

- `model_multimodal.litertlm`
- `tokenizer.json`
- `tokenizer_config.json`
- `chat_template.jinja`
- `vision_encoder_quantized.tflite`
- `vision_adapter_quantized.tflite`
- `embedder_quantized.tflite`
- `model_quantized.tflite`
- `model_multimodal_llm_metadata_multimodal.pb`
- `model_multimodal_export_summary.json`

## Notes

Validated with split-component coverage checks and runtime smoke execution.