{"v":1,"id":"model:hf:handy-computer/canary-qwen-2.5b-gguf","slug":"model-handy-computer-canary-qwen-2-5b-gguf","kind":"model","category":"speech","title":"canary-qwen-2.5b-gguf","summary":"GGUF conversions of nvidia/canary-qwen-2.5b for use with transcribe.cpp.","source":{"provider":"hf","ref":"handy-computer/canary-qwen-2.5b-gguf","url":"https://huggingface.co/handy-computer/canary-qwen-2.5b-gguf","rev":"df576c1641eb59bb66bc3c396bbdd8b0b113825a","fetchedAt":"2026-10-02T21:00:03.612Z","etag":"W/\"1ddf-Yd1B20XE/JO/H+lXPYsInSAACk4\""},"author":{"name":"handy-computer","url":"https://huggingface.co/handy-computer"},"license":{"spdx":"cc-by-4.0","raw":"cc-by-4.0","open":true},"metrics":{"downloads":55992,"likes":1,"takenAt":"2026-10-02T21:00:03.612Z"},"tags":["transcribe.cpp","gguf","asr","speech-to-text","canary","canary-qwen","salm","audio-llm","fastconformer","qwen3","automatic-speech-recognition","en","conversational"],"pipeline":"automatic-speech-recognition","links":{"github":"NVIDIA-NeMo/NeMo"},"updatedAt":"2026-09-15T07:05:30.000Z","collectedAt":"2026-10-02T21:00:03.612Z","review":{"numbers":["55,992 downloads on Hugging Face","1 likes","license cc-by-4.0","1.6 GB for canary-qwen-2.5b-Q4_K_M.gguf"],"log":null},"trust":"unlabeled","health":{"status":"alive","checkedAt":"2026-10-02T21:00:03.612Z","http":200},"description":"# canary-qwen-2.5b: transcribe.cpp GGUF\n\nGGUF conversions of [nvidia/canary-qwen-2.5b](https://huggingface.co/nvidia/canary-qwen-2.5b) for use\nwith [transcribe.cpp](https://github.com/handy-computer/transcribe.cpp).\n\nPorted from upstream commit\n[b1469e1](https://huggingface.co/nvidia/canary-qwen-2.5b/commit/b1469e1),\npinned 2026-05-15.\nValidated against the NeMo SALM 2.7.3 reference at transcribe.cpp commit\n[6f6c699](https://github.com/handy-computer/transcribe.cpp/tree/6f6c699)\non 2026-05-16.\n\nOffline English speech-to-text. NeMo SALM (Speech-Augmented Language\nModel): a FastConformer audio encoder (32 layers, `d_model=1024`) feeds\naudio embeddings into a Qwen3-1.7B causal LM (28 layers,\n`hidden_size=2048`) via audio-token injection at a sentinel position in\nthe prompt. English only. Takes a 16 kHz mono WAV and produces a\ntranscript via greedy decoding.\n\n## Downloads\n\n| Quantization | Download | Size | WER (LibriSpeech test-clean) |\n| --- | --- | ---: | ---: |\n| 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) | 5.08 GB | 1.63% |\n| F16 | [canary-qwen-2.5b-F16.gguf](https://huggingface.co/handy-computer/canary-qwen-2.5b-gguf/resolve/main/canary-qwen-2.5b-F16.gguf) | 5.08 GB | 1.63% |\n| Q8_0 | [canary-qwen-2.5b-Q8_0.gguf](https://huggingface.co/handy-computer/canary-qwen-2.5b-gguf/resolve/main/canary-qwen-2.5b-Q8_0.gguf) | 2.80 GB | 1.63% |\n| Q6_K | [canary-qwen-2.5b-Q6_K.gguf](https://huggingface.co/handy-computer/canary-qwen-2.5b-gguf/resolve/main/canary-qwen-2.5b-Q6_K.gguf) | 2.21 GB | 1.63% |\n| Q5_K_M | [canary-qwen-2.5b-Q5_K_M.gguf](https://huggingface.co/handy-computer/canary-qwen-2.5b-gguf/resolve/main/canary-qwen-2.5b-Q5_K_M.gguf) | 1.98 GB | 1.63% |\n| Q4_K_M | [canary-qwen-2.5b-Q4_K_M.gguf](https://huggingface.co/handy-computer/canary-qwen-2.5b-gguf/resolve/main/canary-qwen-2.5b-Q4_K_M.gguf) | 1.74 GB |…\n\nSource: https://huggingface.co/handy-computer/canary-qwen-2.5b-gguf","install":{"kind":"model","hfId":"handy-computer/canary-qwen-2.5b-gguf","gated":false,"format":"gguf","files":[{"name":"canary-qwen-2.5b-BF16.gguf","size":5076107136,"quant":"BF16","sha256":"5b29334e1d257503b95436608519c39436f0e2056251d17b7c5c001f862c00b8"},{"name":"canary-qwen-2.5b-F16.gguf","size":5076972928,"quant":"F16","sha256":"ae310ac25c1f190b8df4b29a301e936440e1d5306c23d6e5352e29400338ad87"},{"name":"canary-qwen-2.5b-Q4_K_M.gguf","size":1737575808,"quant":"Q4_K_M","sha256":"db5162229d6fa22597d06a613bd9b543eddb3ee02e6afc5e759120fde02bebf7"},{"name":"canary-qwen-2.5b-Q5_K_M.gguf","size":1983729024,"quant":"Q5_K_M","sha256":"d71b1d9ae442dc538665ed4df63a69520eb9be769ec0bc78c7d2f40040928306"},{"name":"canary-qwen-2.5b-Q6_K.gguf","size":2208697728,"quant":"Q6_K","sha256":"25dba2a425113e73d626cc3db6148d925cbd7ead6872b631b7ef4d927788be31"},{"name":"canary-qwen-2.5b-Q8_0.gguf","size":2797548928,"quant":"Q8_0","sha256":"d89aad1285d5bd5aa441c464d3a4cf37bd5474f70705408e71558d8627415b34"}],"totalBytes":18880631552,"suggestedFile":"canary-qwen-2.5b-Q4_K_M.gguf","requirements":{"ramGb":3,"diskBytes":1737575808,"note":"estimate: suggested file size × 1.15 + 0.5 GB; a real measurement comes with lsh models"},"runWith":["whisper.cpp"],"command":"lsh models install hf:handy-computer/canary-qwen-2.5b-gguf"}}