Qwen3-ASR-1.7B GGUF Backup

English | 简体中文 · Also on ModelScope · 42ailab.com

The 1.7B tier of Qwen's speech recognition model: turns recordings into text, larger than the 0.6B tier.

The model was built by the Qwen team at Alibaba and released under Apache-2.0. This repository is neither a new model nor our conversion. The GGUF files were converted by ggml-org and published in ggml-org/Qwen3-ASR-1.7B-GGUF. This repository is a byte-for-byte backup of 2 of those files; nothing was modified.

Why a backup

These files lived only under the converter's own account. If that repository is renamed, deleted, or has its files replaced in place, people who depend on it can no longer download the same bytes.

So we keep the original files, unmodified, on our own ModelScope and Hugging Face accounts. File names, sizes and sha256 digests all match the original.

Use

Packaged for 42model: download Qwen3-ASR-1.7B from the model library in the desktop app, or run 42model download qwen3-asr:1.7b-q8_0.

Files and license

File Role Size sha256
Qwen3-ASR-1.7B-Q8_0.gguf model weights (Q8_0) 2.02 GiB 58e22d0532d4eacaf034cfac17a6fed159f37c41390c710186783be439d1fc57
mmproj-Qwen3-ASR-1.7B-Q8_0.gguf audio projector (Q8_0) 339.2 MiB 46c1d533af3f354ceb37ce855dbceff7da7fa7cf1e6a523df3b13440bd164c0d

The digests can be checked one by one against the file pages of the original ggml-org/Qwen3-ASR-1.7B-GGUF repository.

Weights are under the upstream Apache-2.0 license; the full text is in this repository's LICENSE (taken from Qwen's official GitHub repository QwenLM/Qwen3-ASR; the official model repository ships no separate LICENSE file, and its card is tagged apache-2.0). Credit for the GGUF conversion goes to ggml-org.

Downloads last month
156
GGUF
Model size
2B params
Architecture
qwen3vl
Hardware compatibility
Log In to add your hardware

8-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support