Canary-Qwen 2.5B
ASR and LLM hybrid that transcribes audio then reasons over the text.
About
Canary-Qwen 2.5B is NVIDIA's hybrid speech model released in July 2025, combining a FastConformer speech encoder with a Qwen language model decoder. It works in two modes: an ASR mode that transcribes English speech with punctuation and capitalization, and an LLM mode that keeps the language model's full skills to summarize transcripts or answer questions about them.
With 2.5 billion parameters it topped the Hugging Face OpenASR leaderboard at release with a 5.63% word error rate, and it is licensed under CC-BY-4.0 for commercial use.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 7/10
Scored 7/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 107; Trending score 68/100 from researcher assessment.