canary-qwen-2.5b-gguf

Home/Models/handy-computer/canary-qwen-2.5b-gguf

handy-computer/canary-qwen-2.5b-gguf

kind: gguf
quant: gguf (16-bit)
family: qwen
modality: automatic-speech-recognition
root model: nvidia/canary-qwen-2.5b
provider: handy-computer
license: cc-by-4.0
language: en
context: —
library: transcribe.cpp
vRAM (GB): —
size on disk: —
downloads: 22,112
likes: 1
created: —
last modified: 2026-06-28T13:16:53+00:00 (2026)
huggingface: https://huggingface.co/handy-computer/canary-qwen-2.5b-gguf
tags: transcribe.cppggufasrspeech-to-textcanarycanary-qwensalmaudio-llmfastconformerqwen3automatic-speech-recognitionenarxiv:1706.03762arxiv:2503.05931arxiv:2505.13404arxiv:2310.09424arxiv:2104.02821base_model:nvidia/canary-qwen-2.5bbase_model:quantized:nvidia/canary-qwen-2.5blicense:cc-by-4.0region:usconversational

benchmark	category	score	context	conf	evidence
WER	speech-asr	16.71	% WER	medium	`\| % WER \| 16.71 \| 13.85 \| 17.71 \| 29.46 \|`
WER Earnings-22	speech-asr	10.42		low	`\| 2.5.0 \| Canary-Qwen-2.5B \| 418 \| 5.63 \| 10.18 \| 9.41 \| 1.60 \| 3.10 \| 10.42 \| 1.90 \| 2.72 \| 5.66 \|`
WER AMI	speech-asr	10.18		low	`\| 2.5.0 \| Canary-Qwen-2.5B \| 418 \| 5.63 \| 10.18 \| 9.41 \| 1.60 \| 3.10 \| 10.42 \| 1.90 \| 2.72 \| 5.66 \|`
WER GigaSpeech	speech-asr	9.41		low	`\| 2.5.0 \| Canary-Qwen-2.5B \| 418 \| 5.63 \| 10.18 \| 9.41 \| 1.60 \| 3.10 \| 10.42 \| 1.90 \| 2.72 \| 5.66 \|`
WER VoxPopuli	speech-asr	5.66		low	`\| 2.5.0 \| Canary-Qwen-2.5B \| 418 \| 5.63 \| 10.18 \| 9.41 \| 1.60 \| 3.10 \| 10.42 \| 1.90 \| 2.72 \| 5.66 \|`
WER TED-LIUM	speech-asr	2.72		low	`\| 2.5.0 \| Canary-Qwen-2.5B \| 418 \| 5.63 \| 10.18 \| 9.41 \| 1.60 \| 3.10 \| 10.42 \| 1.90 \| 2.72 \| 5.66 \|`
WER	speech-asr	1.63		low	`\| BF16 \| [canary-qwen-2.5b-BF16.gguf](https://huggingface.co/handy-computer/canary-qwen-2.5b-gguf/resolve/main/canary-qwen-2.5b-BF16.gguf) \| 4.73 GB \| 1.63% \|`

none scraped.