| 1 | granite-speech-4.1-2b | ibm-granite/granite-4.0-1b-base | ibm-granite | 24.86 | medium | VoxPopuli |
| 2 | granite-speech-4.1-2b-gguf | ibm-granite/granite-speech-4.1-2b | handy-computer | 24.86 | medium | VoxPopuli |
| 3 | granite-speech-4.1-2b-gguf | ibm-granite/granite-speech-4.1-2b | memoravox | 24.86 | medium | VoxPopuli |
| 4 | whisper-small-cv11-french | — | bofenghuang | 15.7 | low | |
| 5 | bp500-base100k_voxpopuli | — | lgris | 14.2 | medium | bp\_500-base100k_voxpopuli (demonstration below) |
| 6 | moonshine-streaming-tiny | UsefulSensors/moonshine-streaming-tiny | UsefulSensors | 14.02 | medium | VoxPopuli |
| 7 | moonshine-streaming-medium-gguf | UsefulSensors/moonshine-streaming-medium | handy-computer | 14.02 | medium | VoxPopuli |
| 8 | moonshine-streaming-small-gguf | UsefulSensors/moonshine-streaming-small | handy-computer | 14.02 | medium | VoxPopuli |
| 9 | moonshine-streaming-medium | UsefulSensors/moonshine-streaming-medium | UsefulSensors | 14.02 | medium | VoxPopuli |
| 10 | moonshine-streaming-small | UsefulSensors/moonshine-streaming-small | UsefulSensors | 14.02 | medium | VoxPopuli |
| 11 | moonshine-streaming-tiny-gguf | UsefulSensors/moonshine-streaming-tiny | handy-computer | 14.02 | medium | VoxPopuli |
| 12 | stt_it_fastconformer_hybrid_large_pc | nvidia/stt_it_fastconformer_hybrid_large_pc | nvidia | 13.02 | low | |
| 13 | hviske-v5 | — | syvai | 11.8 | medium | VoxPopuli |
| 14 | stt_pl_fastconformer_hybrid_large_pc | nvidia/stt_pl_fastconformer_hybrid_large_pc | nvidia | 11.29 | low | |
| 15 | stt_de_conformer_transducer_large | nvidia/stt_de_conformer_transducer_large | nvidia | 11.04 | low | |
| 16 | stt_de_fastconformer_hybrid_large_pc | nvidia/stt_de_fastconformer_hybrid_large_pc | nvidia | 11.03 | low | |
| 17 | nemotron-3.5-asr-streaming-0.6b-onnx-int4 | nvidia/nemotron-3.5-asr-streaming-0.6b | onnx-community | 9.96 | high | |
| 18 | bp500-base100k_voxpopuli | — | lgris | 9.9 | medium | bp\_500-base100k_voxpopuli + 4-gram (demonstration below) |
| 19 | parakeet_realtime_eou_120m-v1 | nvidia/parakeet_realtime_eou_120m-v1 | nvidia | 9.07 | low | |
| 20 | CrisperWhisper | openai/whisper-large-v3 | nyralabs | 8.61 | medium | [VoxPopuli](https://huggingface.co/datasets/facebook/voxpopuli) |
| 21 | CrisperWhisper | openai/whisper-large-v3 | unsloth | 8.61 | medium | [VoxPopuli](https://huggingface.co/datasets/facebook/voxpopuli) |
| 22 | kotoba-whisper-bilingual-v1.0 | — | kotoba-tech | 8.3 | low | |
| 23 | VibeVoice-ASR-HF | microsoft/VibeVoice-ASR-HF | microsoft | 8.01 | medium | voxpopuli_test |
| 24 | wav2vec2-xlsr-300m-finnish-lm | — | Finnish-NLP | 7.97 | medium | [VoxPopuli Finnish](https://github.com/facebookresearch/voxpopuli) |
| 25 | wav2vec2-xlsr-1b-finnish-lm-v2 | — | Finnish-NLP | 7.97 | medium | [VoxPopuli Finnish](https://github.com/facebookresearch/voxpopuli) |
| 26 | wav2vec2-large-uralic-voxpopuli-v2-finnish | — | Finnish-NLP | 7.94 | medium | [VoxPopuli Finnish](https://github.com/facebookresearch/voxpopuli) |
| 27 | nemotron-speech-streaming-en-0.6b | nvidia/nemotron-speech-streaming-en-0.6b | nvidia | 7.91 | low | |
| 28 | nemotron-speech-streaming-en-0.6b-gguf | nvidia/nemotron-speech-streaming-en-0.6b | handy-computer | 7.91 | low | |
| 29 | Nemotron-Speech-Streaming-0.6B-CoreML-INT8 | nvidia/nemotron-speech-streaming-en-0.6b | aufklarer | 7.91 | medium | VoxPopuli |
| 30 | nemotron-speech-streaming-en-0.6b-gguf | nvidia/nemotron-speech-streaming-en-0.6b | memoravox | 7.91 | low | |
| 31 | multitalker-parakeet-streaming-0.6b-v1-gguf | nvidia/multitalker-parakeet-streaming-0.6b-v1 | handy-computer | 7.57 | low | |
| 32 | multitalker-parakeet-streaming-0.6b-v1 | nvidia/multitalker-parakeet-streaming-0.6b-v1 | nvidia | 7.57 | low | |
| 33 | thewhisper-large-v3-turbo | openai/whisper-large-v3-turbo | TheStageAI | 7.37 | medium | **VoxPopuli** |
| 34 | LFM2.5-Audio-1.5B | LiquidAI/LFM2-1.2B | LiquidAI | 7.13 | high | Open |
| 35 | LFM2.5-Audio-1.5B-GGUF | LiquidAI/LFM2-1.2B | Mungert | 7.13 | high | Open |
| 36 | parakeet-tdt_ctc-110m-gguf | nvidia/parakeet-tdt_ctc-110m | handy-computer | 6.91 | low | |
| 37 | parakeet-tdt_ctc-110m | nvidia/parakeet-tdt_ctc-110m | nvidia | 6.91 | low | |
| 38 | asr-conformer-loquacious | — | speechbrain | 6.9 | low | |
| 39 | Qwen3-ASR-0.6B-hf | Qwen/Qwen3-ASR-0.6B-hf | Qwen | 6.8 | high | |
| 40 | Qwen3-ASR-0.6B-hf | Qwen/Qwen3-ASR-0.6B | bezzam | 6.8 | high | |
| 41 | ARK-ASR-0.6B | — | Audio8 | 6.63 | high | |
| 42 | Ming-Lite-Omni-1.5 | inclusionAI/Ming-Lite-Omni | inclusionAI | 6.43 | high | Average on All/Open-source Benchmarks(↓) |
| 43 | canary-180m-flash-gguf | nvidia/canary-180m-flash | handy-computer | 6.35 | low | |
| 44 | canary-180m-flash | nvidia/canary-180m-flash | nvidia | 6.35 | low | |
| 45 | canary-1b-v2-gguf | nvidia/canary-1b-v2 | handy-computer | 6.25 | low | |
| 46 | canary-1b-v2 | nvidia/canary-1b-v2 | nvidia | 6.25 | low | |
| 47 | canary-1b-v2-gguf | nvidia/canary-1b-v2 | memoravox | 6.25 | low | |
| 48 | parakeet-tdt_ctc-1.1b-gguf | nvidia/parakeet-tdt_ctc-1.1b | handy-computer | 6.19 | low | |
| 49 | parakeet-tdt_ctc-1.1b | nvidia/parakeet-tdt_ctc-1.1b | nvidia | 6.19 | low | |
| 50 | parakeet-tdt-0.6b-v3-gguf | nvidia/parakeet-tdt-0.6b-v3 | handy-computer | 6.14 | low | |
| 51 | parakeet-tdt-0.6b-v3 | nvidia/parakeet-tdt-0.6b-v3 | nvidia | 6.14 | low | |
| 52 | higgs-audio-v3-stt-v2 | bosonai/higgs-audio-v3-stt | bosonai | 6.07 | medium | VoxPopuli |
| 53 | Qwen3-ASR-1.7B-hf | Qwen/Qwen3-ASR-1.7B-hf | Qwen | 5.99 | high | |
| 54 | Qwen3-ASR-1.7B-hf | Qwen/Qwen3-ASR-1.7B | bezzam | 5.99 | high | |
| 55 | parakeet-tdt-0.6b-v2 | nvidia/parakeet-tdt-0.6b-v2 | nvidia | 5.95 | high | |
| 56 | parakeet-tdt-0.6b-v2-gguf | nvidia/parakeet-tdt-0.6b-v2 | handy-computer | 5.95 | high | |
| 57 | higgs-audio-v3-8b-stt-v2 | bosonai/higgs-audio-v3-8b | bosonai | 5.92 | medium | VoxPopuli |
| 58 | granite-speech-4.1-2b-plus-gguf | ibm-granite/granite-speech-4.1-2b-plus | handy-computer | 5.9 | low | |
| 59 | granite-speech-4.1-2b-plus-gguf | ibm-granite/granite-speech-4.1-2b-plus | memoravox | 5.9 | low | |
| 60 | stt_fr_fastconformer_hybrid_large_pc | nvidia/stt_fr_fastconformer_hybrid_large_pc | nvidia | 5.86 | low | |
| 61 | wav2vec2-xlsr-300m-finnish-lm | — | Finnish-NLP | 5.85 | medium | Finnish-NLP/wav2vec2-base-fi-voxpopuli-v2-finetuned |
| 62 | wav2vec2-large-uralic-voxpopuli-v2-finnish | — | Finnish-NLP | 5.85 | medium | Finnish-NLP/wav2vec2-base-fi-voxpopuli-v2-finetuned |
| 63 | wav2vec2-xlsr-1b-finnish-lm-v2 | — | Finnish-NLP | 5.85 | medium | Finnish-NLP/wav2vec2-base-fi-voxpopuli-v2-finetuned |
| 64 | granite-4.0-1b-speech | ibm-granite/granite-4.0-1b-base | ibm-granite | 5.84 | high | |
| 65 | granite-4.0-1b-speech-gguf | ibm-granite/granite-4.0-1b-speech | handy-computer | 5.84 | low | |
| 66 | granite-4.0-1b-speech-ONNX | ibm-granite/granite-4.0-1b-speech | onnx-community | 5.84 | low | |
| 67 | granite-speech-4.1-2b-nar-gguf | ibm-granite/granite-speech-4.1-2b-nar | handy-computer | 5.83 | medium | VoxPopuli |
| 68 | granite-speech-4.1-2b-nar-gguf | ibm-granite/granite-speech-4.1-2b-nar | memoravox | 5.83 | medium | VoxPopuli |
| 69 | Ming-Lite-Omni | inclusionAI/Ling-lite | inclusionAI | 5.8 | high | |
| 70 | granite-speech-4.1-2b-plus | ibm-granite/granite-4.0-1b-base | ibm-granite | 5.7 | high | |
| 71 | canary-qwen-2.5b | nvidia/canary-1b-flash | nvidia | 5.66 | low | |
| 72 | canary-qwen-2.5b-gguf | nvidia/canary-qwen-2.5b | handy-computer | 5.66 | low | |
| 73 | canary-1b-flash-gguf | nvidia/canary-1b-flash | handy-computer | 5.63 | low | |
| 74 | canary-1b-flash | nvidia/canary-1b-flash | nvidia | 5.63 | low | |
| 75 | higgs-audio-v3-8b-stt | bosonai/higgs-audio-v3-8b | bosonai | 5.63 | medium | VoxPopuli |
| 76 | canary-1b-flash-gguf | nvidia/canary-1b-flash | memoravox | 5.63 | low | |
| 77 | Zipformer-cr-ctc-transducer-XL-290M | — | soundsgoodai | 5.6 | medium | VoxPopuli |
| 78 | ARK-ASR-3B | AutoArk-AI/ARK-ASR-3B | Audio8 | 5.47 | high | |
| 79 | MOSS-Transcribe-preview-2B | OpenMOSS-Team/MOSS-Transcribe-preview-2B | OpenMOSS-Team | 5.39 | medium | VoxPopuli |
| 80 | VibeVoice-ASR-BitNet | microsoft/VibeVoice-ASR-BitNet | microsoft | 4.92 | medium | VoxPopuli |
| 81 | stt_es_fastconformer_hybrid_large_pc | nvidia/stt_es_fastconformer_hybrid_large_pc | nvidia | 4.62 | low | |
| 82 | stt_en_fastconformer_hybrid_large_pc | nvidia/stt_en_fastconformer_hybrid_large_pc | nvidia | 4.22 | low | |
| 83 | wav2vec2-xlsr-300m-finnish-lm | — | Finnish-NLP | 4.13 | medium | Finnish-NLP/wav2vec2-large-uralic-voxpopuli-v2-finnish |
| 84 | wav2vec2-large-uralic-voxpopuli-v2-finnish | — | Finnish-NLP | 4.13 | medium | Finnish-NLP/wav2vec2-large-uralic-voxpopuli-v2-finnish |
| 85 | wav2vec2-xlsr-1b-finnish-lm-v2 | — | Finnish-NLP | 4.13 | medium | Finnish-NLP/wav2vec2-large-uralic-voxpopuli-v2-finnish |
| 86 | parakeet-rnnt-1.1b-gguf | nvidia/parakeet-rnnt-1.1b | handy-computer | 3.92 | low | |
| 87 | parakeet-rnnt-1.1b | nvidia/parakeet-rnnt-1.1b | nvidia | 3.92 | low | |
| 88 | parakeet-rnnt-0.6b | nvidia/parakeet-rnnt-0.6b | nvidia | 3.86 | low | |
| 89 | parakeet-rnnt-0.6b-gguf | nvidia/parakeet-rnnt-0.6b | handy-computer | 3.86 | low | |
| 90 | parakeet-rnnt-0.6b-gguf | nvidia/parakeet-rnnt-0.6b | memoravox | 3.86 | low | |
| 91 | parakeet-ctc-0.6b-gguf | nvidia/parakeet-ctc-0.6b | handy-computer | 3.78 | low | |
| 92 | parakeet-ctc-0.6b | nvidia/parakeet-ctc-0.6b | nvidia | 3.78 | low | |
| 93 | parakeet-tdt-1.1b-gguf | nvidia/parakeet-tdt-1.1b | handy-computer | 3.56 | low | |
| 94 | parakeet-tdt-1.1b | nvidia/parakeet-tdt-1.1b | nvidia | 3.56 | low | |
| 95 | parakeet-ctc-1.1b | nvidia/parakeet-ctc-1.1b | nvidia | 3.54 | low | |
| 96 | parakeet-ctc-1.1b-gguf | nvidia/parakeet-ctc-1.1b | handy-computer | 3.54 | low | |