Open ASR · AMI-Cleaned WER ↗English short-form; normalization unversioned
7.91% WERnvidia/canary-qwen-2.5b
nvidiaDiscovered Oct 17, 2026
nvidia/canary-qwen-2.5b
2.5B parameters · Encoder: Conformer-based · Decoder: Qwen2
Measured performance
All leaderboards ↗Benchmark results
Word error rate · lower is better
Open ASR · Earnings22-Cleaned-AA-chunked WER ↗English short-form; normalization unversioned
6.41% WERnvidia/canary-qwen-2.5b
Open ASR · Gigaspeech-Cleaned WER ↗English short-form; normalization unversioned
7.8% WERnvidia/canary-qwen-2.5b
Open ASR · LS Clean WER ↗English short-form; normalization unversioned
1.23% WERnvidia/canary-qwen-2.5b
Open ASR · LS Other WER ↗English short-form; normalization unversioned
2.63% WERnvidia/canary-qwen-2.5b
Open ASR · SPGISpeech WER ↗English short-form; normalization unversioned
1.7% WERnvidia/canary-qwen-2.5b
Open ASR · URGENT2024 WER ↗English short-form; normalization unversioned
13.02% WERnvidia/canary-qwen-2.5b
Open ASR · Voice Arena Monsoon WER ↗English short-form; normalization unversioned
3.6% WERnvidia/canary-qwen-2.5b
Open ASR · Voxpopuli-AA-Cleaned WER ↗English short-form; normalization unversioned
4.14% WERnvidia/canary-qwen-2.5b
Access & cost
0 quotesPrices across providers
No listed provider price
Model specifications
- Developer
- nvidia
- Modality
- Audio
- Context
- Not reported
- Maximum output
- Not reported
- Weights
- Not reported
- First discovered
- Oct 17, 2026
More from nvidia
View all ↗nvidiaListed 22 days ago
Nemotron-3-Diarization
Nemotron 3 Diarization is an open-weight streaming Sortformer model, engineered for low-latency speaker diarization across up to 8 speakers, ordered by arrival time. It powers meeting transcription, call…
No reported score
No reported score
nvidiaListed Jun 3
Nemotron-3.5-ASR-Streaming-Multilingual-0.6b
Nemotron 3.5 ASR Streaming Multilingual is an open 0.6B-parameter prompt-conditioned cache-aware FastConformer-RNNT model, engineered for low-latency streaming transcription across 40+ languages. It powers…
0.6BAudio inputOpen weightsNo reported score
No reported score
nvidiaListed Jun 1
Nemotron Asr Multilingual
Nemotron-ASR-Streaming is a multi lingual, streaming Automatic Speech Recognition (ASR) engineered to deliver high-quality multi lingual transcription across both low-latency streaming and high-throughput…
No reported score
No reported score
nvidiaListed Apr 27
Nemotron 3 Nano Omni
Audio reasoning variant of NVIDIA's Nemotron 3 Nano Omni. 30B A3B hybrid Transformer-Mamba MoE - accepts audio plus a prompt and returns text.
No reported score
No reported score
Price not listed