HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks

Qwen3-ASR-1.7B-Swift

YoozLabs/Qwen3-ASR-1.7B-Swift

kind
mlx
quant
mlx
family
qwen
modality
automatic-speech-recognition
root model
mlx-community/Qwen3-ASR-1.7B-8bit
provider
YoozLabs
license
apache-2.0
language
en, ar, fa, zh
context
—
library
mlx
vRAM (GB)
—
size on disk
—
downloads
137
likes
0
created
—
last modified
2026-07-20T04:19:41+00:00 (2026)
huggingface
https://huggingface.co/YoozLabs/Qwen3-ASR-1.7B-Swift
tags
mlxsafetensorsqwen3_asrasrautomatic-speech-recognitionswiftqwen3multilingualenarfazhbase_model:mlx-community/Qwen3-ASR-1.7B-8bitbase_model:quantized:mlx-community/Qwen3-ASR-1.7B-8bitlicense:apache-2.08-bitregion:us

Benchmark scores (1)

benchmarkcategoryscorecontextconfevidence
WER LibriSpeechspeech-asr6.3LibriSpeech-style Englishmedium| LibriSpeech-style English | WER | **6.3%** | Parakeet TDT on the same set is 6.9% |

Discussions (0)

none scraped.

Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.