| 01 | Stt_en_fastconformer_ctc_large | unverified | 6399.25031 | 2023 | Paper ↗Source ↗ | Looks wrong? |
| 02 | Stt_en_conformer_ctc_small | unverified | 5686.896503 | 2020 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 03 | Parakeet-tdt_ctc-110m | unverified | 5345.14 | 2023 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 04 | NVIDIA NeMo Conformer-CTC Large RTFx (real-time factor, higher=faster). nvidia/stt_en_conformer_ctc_large. 120M Conformer CTC. Fast inference (RTFx 4295). | verified | 4295.01 | 2023 | Source ↗ | Looks wrong? |
| 05 | Stt_en_conformer_ctc_large | unverified | 4295.006653 | 2020 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 06 | Parakeet-ctc-0.6b | unverified | 4281.529811 | 2023 | Paper ↗Source ↗ | Looks wrong? |
| 07 | Stt_en_fastconformer_transducer_large | unverified | 4097.432343 | 2023 | Paper ↗Source ↗ | Looks wrong? |
| 08 | Parakeet-tdt-0.6b-v2 | unverified | 3386.02 | 2023 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 09 | Parakeet TDT 0.6B v2 RTFx (real-time factor, higher=faster). Rank #10. nvidia/parakeet-tdt-0.6b-v2. RTFx 3386 (fastest open model). | verified | 3386.02 | 2024 | Source ↗ | Looks wrong? |
| 10 | Parakeet TDT 0.6B v3 RTFx (real-time factor, higher=faster). Rank #12. nvidia/parakeet-tdt-0.6b-v3. | verified | 3332.74 | 2025 | Source ↗ | Looks wrong? |
| 11 | Parakeet-rnnt-0.6b | unverified | 2815.724575 | 2023 | Paper ↗Source ↗ | Looks wrong? |
| 12 | Parakeet-TDT-1.1B RTFx (real-time factor, higher=faster). nvidia/parakeet-tdt-1.1b. RTFx 2390. | verified | 2390.61 | 2024 | Source ↗ | Looks wrong? |
| 13 | Canary-180M-Flash RTFx (real-time factor, higher=faster). nvidia/canary-180m-flash. 182M params, RTFx 1233. | verified | 1233.58 | 2025 | Source ↗ | Looks wrong? |
| 14 | Canary-1B-Flash RTFx (real-time factor, higher=faster). Rank #13. nvidia/canary-1b-flash. RTFx 1045. | verified | 1045.75 | 2025 | Source ↗ | Looks wrong? |
| 15 | Moonshine-streaming-tiny | unverified | 847.2 | 2026 | Paper ↗ | Looks wrong? |
| 16 | Moonshine-tiny | unverified | 753.06 | 2024 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 17 | Canary 1B v2 RTFx (real-time factor, higher=faster). nvidia/canary-1b-v2. | verified | 749 | 2024 | Source ↗ | Looks wrong? |
| 18 | Wav2vec2-base-960h | unverified | 686.002907 | 2020 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 19 | Data2vec-audio-base-960h | unverified | 648.138532 | 2022 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 20 | Wav2vec2-conformer-rope-large-960h-ft | unverified | 607.869462 | 2020 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 21 | Moonshine-streaming-small | unverified | 566.33 | 2026 | Paper ↗Source ↗ | Looks wrong? |
| 22 | Moonshine-base | unverified | 565.97 | 2024 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 23 | Cohere Transcribe (Mar 2026) | unverified | 524.88 | 2026 | Paper ↗ | Looks wrong? |
| 24 | Wav2vec2-conformer-rel-pos-large-960h-ft | unverified | 522.456837 | 2020 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 25 | wav2vec 2.0 Large (960h) | unverified | 516.579659 | 2020 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 26 | Wav2vec2-large-960h-lv60-self | unverified | 509.320417 | 2020 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 27 | Wav2vec2-large-robust-ft-libri-960h | unverified | 503.808561 | 2021 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 28 | Owsm_ctc_v3.1_1B | unverified | 502.02 | 2024 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 29 | Hubert-large-ls960-ft | unverified | 495.862704 | 2021 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 30 | Data2vec-audio-large-960h | unverified | 470.154204 | 2022 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 31 | Asr-wav2vec2-librispeech | unverified | 451.181976 | 2021 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 32 | Moonshine Streaming Medium | unverified | 448.15 | 2026 | Paper ↗Source ↗ | Looks wrong? |
| 33 | Canary-Qwen-2.5B RTFx (real-time factor, higher=faster). Rank #4. nvidia/canary-qwen-2.5b. FastConformer encoder + Qwen2 LM. | verified | 418.28 | 2025 | Source ↗ | Looks wrong? |
| 34 | Hubert-xlarge-ls960-ft | unverified | 361.317654 | 2021 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 35 | Whisper-tiny.en | unverified | 348.123935 | 2022 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 36 | Distil-small.en | unverified | 331.893486 | 2023 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 37 | Whisper-base.en | unverified | 320.673885 | 2022 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 38 | Granite 4.0 1B Speech RTFx (real-time factor, higher=faster). Rank #3. ibm-granite/granite-4.0-1b-speech. 1B open model. | verified | 280.02 | 2025 | Source ↗ | Looks wrong? |
| 39 | Distil-medium.en | unverified | 279.733104 | 2023 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 40 | Granite Speech 3.3 2B RTFx (real-time factor, higher=faster). Rank #8. ibm-granite/granite-speech-3.3-2b. | verified | 270.57 | 2025 | Source ↗ | Looks wrong? |
| 41 | Whisper-small.en | unverified | 268.914874 | 2022 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 42 | Canary-1B RTFx (real-time factor, higher=faster). nvidia/canary-1b. FastConformer encoder + T5 decoder. | verified | 235.34 | 2024 | Source ↗ | Looks wrong? |
| 43 | Mms-1b-fl102 | unverified | 234.423174 | 2023 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 44 | Granite Speech 4.1 2B | unverified | 231.29 | 2025 | Paper ↗Source ↗ | Looks wrong? |
| 45 | Mms-1b-all | unverified | 230.794251 | 2023 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 46 | Distil-large-v3 | unverified | 214.421431 | 2023 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 47 | Distil-Whisper Large v3 RTFx (real-time factor, higher=faster). distil-whisper/distil-large-v3. | verified | 214.42 | 2024 | Source ↗ | Looks wrong? |
| 48 | Distil-large-v2 | unverified | 202.946441 | 2023 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 49 | Distil-Whisper Large v3.5 RTFx (real-time factor, higher=faster). distil-whisper/distil-large-v3.5. 756M params. | verified | 202.03 | 2024 | Source ↗ | Looks wrong? |
| 50 | Whisper Large v3 Turbo | unverified | 200.19 | 2022 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 51 | Lite-whisper-large-v3-turbo-acc | unverified | 191.71 | 2025 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 52 | Whisper-medium.en | unverified | 182.12916 | 2022 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 53 | Qwen3-ASR-0.6B RTFx (real-time factor, higher=faster). Rank #15 (tied). Qwen/Qwen3-ASR-0.6B. Smallest Qwen3-ASR variant. | verified | 166.23 | 2025 | Source ↗ | Looks wrong? |
| 54 | Phi-4 Multimodal Instruct RTFx (real-time factor, higher=faster). Rank #9. microsoft/Phi-4-multimodal-instruct. 6B param open model. | verified | 151.1 | 2025 | Source ↗ | Looks wrong? |
| 55 | Qwen3-ASR-1.7B | unverified | 147.93 | 2026 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 56 | Whisper Large v3 RTFx (real-time factor, higher=faster). openai/whisper-large-v3. Baseline Whisper entry on the HF Open ASR Leaderboard. | verified | 145.51 | 2023 | Source ↗ | Looks wrong? |
| 57 | Granite Speech 3.3 8B RTFx (real-time factor, higher=faster). Rank #5. ibm-granite/granite-speech-3.3-8b. 8B open model. | verified | 145.42 | 2025 | Source ↗ | Looks wrong? |
| 58 | GLM-ASR-Nano-2512 RTFx (real-time factor, higher=faster). zai-org/GLM-ASR-Nano-2512. GLM4 2B + audio encoder. | verified | 145.28 | 2024 | Source ↗ | Looks wrong? |
| 59 | Whisper Large v2 | unverified | 144.452102 | 2022 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 60 | Whisper Large | unverified | 143.756319 | 2022 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 61 | Lite-whisper-large-v3-fast | unverified | 120.76 | 2025 | Paper ↗Code ↗Source ↗ | Looks wrong? |
| 62 | Voxtral Mini 3B RTFx (real-time factor, higher=faster). mistralai/Voxtral-Mini-3B-2507. 5B effective params. | verified | 109.86 | 2025 | Source ↗ | Looks wrong? |
| 63 | Voxtral-Mini-4B-Realtime-2602 | unverified | 93.32 | 2026 | Paper ↗Source ↗ | Looks wrong? |
| 64 | Kyutai STT 2.6B (EN) RTFx (real-time factor, higher=faster). Rank #14. kyutai/stt-2.6b-en. Streaming ASR based on Moshi architecture. | verified | 88.37 | 2024 | Source ↗ | Looks wrong? |
| 65 | CrisperWhisper RTFx (real-time factor, higher=faster). nyrahealth/CrisperWhisper. Whisper fine-tune with precise word boundaries. | verified | 84.05 | 2024 | Source ↗ | Looks wrong? |
| 66 | SYMPHONY-ASR | unverified | 77.56 | 2026 | Paper ↗ | Looks wrong? |
| 67 | Voxtral-Small-24B-2507 RTFx (real-time factor, higher=faster). mistralai/Voxtral-Small-24B-2507. Mistral large multimodal model. | verified | 54.09 | 2025 | Source ↗ | Looks wrong? |
| 68 | VibeVoice-ASR-HF | unverified | 51.8 | 2026 | Paper ↗Source ↗ | Looks wrong? |
| 69 | Asr-conformer-loquacious | unverified | 42.16 | 2025 | Paper ↗ | Looks wrong? |