skip to content
gigarouter gigarouter
rankings / transcribe-speech

The best speech-to-text models

92 models & services · 0 callable here now

Ranked by benchmark score per dollar (quality floor applied). Scores: Open ASR Leaderboard — average word error rate across 8 English test sets - lower is better. Fetched 2026-07-04. Prices are our live per-call rates; ~ marks an estimate until the model is onboarded.

sorted by value · sort by score

#modelscorepriceparamsstatus
1AutoArk-AI/ARK-ASR-3B best value4.8 (#1)~$0.0034 / minute4063.4Mcoming soon
2OpenMOSS-Team/MOSS-Transcribe-preview-2B4.9 (#2)~$0.0034 / minute2418.8Mcoming soon
3AutoArk-AI/ARK-ASR-0.6B5.5 (#4)~$0.0034 / minute1299.5Mcoming soon
4Qwen/Qwen3-ASR-1.7B5.8 (#5)~$0.0034 / minute2349.2Mcoming soon
5microsoft/Phi-4-multimodal-instruct6 (#6)~$0.0034 / minute5574.5Mcoming soon
6nvidia/canary-1b-flash6.4 (#9)~$0.0034 / minute811Mcoming soon
7kyutai/stt-2.6b-en6.4 (#10)~$0.0034 / minute2617.1Mcoming soon
8Qwen/Qwen3-ASR-0.6B6.4 (#11)~$0.0034 / minute938Mcoming soon
9UsefulSensors/moonshine-streaming-medium6.7 (#13)~$0.0034 / minute265.9Mcoming soon
10zai-org/GLM-ASR-Nano-25127 (#16)~$0.0034 / minute2257.8Mcoming soon
11nvidia/parakeet-rnnt-1.1b7.1 (#19)~$0.0034 / minute1070.5Mcoming soon
12distil-whisper/distil-large-v3.57.2 (#21)~$0.0034 / minute756.4Mcoming soon
13nvidia/parakeet-ctc-1.1b7.4 (#22)~$0.0034 / minute1062.6Mcoming soon
14nvidia/parakeet-rnnt-0.6b7.5 (#26)~$0.0034 / minute616.7Mcoming soon
15nvidia/parakeet-ctc-0.6b7.7 (#27)~$0.0034 / minute608.8Mcoming soon
16openai/whisper-large-v27.8 (#29)~$0.0034 / minute1543.3Mcoming soon
17UsefulSensors/moonshine-streaming-small7.8 (#30)~$0.0034 / minute140.1Mcoming soon
18openai/whisper-large7.9 (#31)~$0.0034 / minute1543.3Mcoming soon
19openai/whisper-medium.en8.1 (#32)~$0.0034 / minute763.9Mcoming soon
20openai/whisper-small.en8.6 (#36)~$0.0034 / minute241.7Mcoming soon
21CohereLabs/cohere-transcribe-03-20265.4 (#3)-2065.8Mnot hosted
22nvidia/parakeet-tdt-0.6b-v26.1 (#7)--not hosted
23nvidia/parakeet-tdt-0.6b-v36.3 (#8)-627.1Mnot hosted
24nvidia/canary-1b6.5 (#12)--not hosted
25soundsgoodai/Zipformer-transducer-XL-290M6.7 (#14)--not hosted
26nvidia/parakeet-tdt-1.1b7 (#15)--not hosted
27mistralai/Voxtral-Mini-3B-25077.1 (#17)-4676.3Mcoming soon
28nvidia/canary-180m-flash7.1 (#18)--not hosted
29nvidia/canary-1b-v27.2 (#20)--not hosted
30espnet/owsm_ctc_v4_1B7.4 (#23)--not hosted
31openai/whisper-large-v37.4 (#24)-1543.5Mnot hosted
32nvidia/parakeet-tdt_ctc-110m7.5 (#25)--not hosted
33microsoft/VibeVoice-ASR-HF7.8 (#28)-8330.3Mcoming soon
34espnet/owsm_ctc_v3.1_1B8.1 (#33)--not hosted
35nvidia/stt_en_conformer_ctc_large8.3 (#34)--not hosted
36speechbrain/asr-conformer-loquacious8.5 (#35)--not hosted
37abr-ai/niagara-38m-batch.en8.9 (#37)--not hosted
38nvidia/stt_en_fastconformer_ctc_large9 (#38)--not hosted
39nvidia/stt_en_fastconformer_transducer_large9.1 (#39)--not hosted
40UsefulSensors/moonshine-base10 (#40)~$0.0034 / minute61.5Mcoming soon
41openai/whisper-base.en10.3 (#41)~$0.0034 / minute72.6Mcoming soon
42abr-ai/niagara-19m-batch.en10.5 (#42)--not hosted
43nvidia/stt_en_conformer_ctc_small11.2 (#43)--not hosted
44UsefulSensors/moonshine-streaming-tiny12 (#44)-44.1Mnot hosted
45UsefulSensors/moonshine-tiny12.7 (#45)-27.1Mnot hosted
46openai/whisper-tiny.en12.8 (#46)-37.8Mnot hosted
47speechbrain/asr-wav2vec2-librispeech14.3 (#47)--not hosted
48facebook/wav2vec2-large-960h-lv60-self21.3 (#48)--not hosted
49facebook/mms-1b-all22.5 (#49)-964.8Mnot hosted
50facebook/hubert-xlarge-ls960-ft22.5 (#50)-962.5Mnot hosted
51facebook/hubert-large-ls960-ft22.7 (#51)--not hosted
52facebook/wav2vec2-large-robust-ft-libri-960h22.9 (#52)-315.5Mnot hosted
53facebook/data2vec-audio-large-960h23.2 (#53)--not hosted
54facebook/wav2vec2-conformer-rope-large-960h-ft23.3 (#54)-593.4Mnot hosted
55facebook/wav2vec2-conformer-rel-pos-large-960h-ft23.3 (#55)--not hosted
56facebook/wav2vec2-large-960h26.8 (#56)--not hosted
57facebook/data2vec-audio-base-960h28.3 (#57)--not hosted
58facebook/wav2vec2-base-960h29.4 (#58)-94.4Mnot hosted
59facebook/mms-1b-fl10239.8 (#59)-964.7Mnot hosted
60pyannote/speaker-diarization-3.1---coming soon
61argmaxinc/whisperkit-coreml---coming soon
62openai/whisper-base-~$0.0034 / minute72.6Mcoming soon
63jonatasgrosman/wav2vec2-large-xlsr-53-japanese---coming soon
64jonatasgrosman/wav2vec2-large-xlsr-53-polish---coming soon
65jonatasgrosman/wav2vec2-large-xlsr-53-dutch---coming soon
66indonesian-nlp/wav2vec2-indonesian-javanese-sundanese---coming soon
67pyannote/speaker-diarization-community-1---coming soon
68jonatasgrosman/wav2vec2-large-xlsr-53-arabic---coming soon
69jonatasgrosman/wav2vec2-large-xlsr-53-hungarian---coming soon
70openai/whisper-small-~$0.0034 / minute241.7Mcoming soon
71MahmoudAshraf/mms-300m-1130-forced-aligner-~$0.0034 / minute315.5Mcoming soon
72jonatasgrosman/wav2vec2-large-xlsr-53-portuguese---coming soon
73jonatasgrosman/wav2vec2-large-xlsr-53-russian---coming soon
74gigant/romanian-wav2vec2-~$0.0034 / minute315.5Mcoming soon
75anuragshas/wav2vec2-large-xlsr-53-telugu---coming soon
76jonatasgrosman/wav2vec2-large-xlsr-53-persian---coming soon
77KBLab/wav2vec2-large-voxrex-swedish-~$0.0034 / minute315.5Mcoming soon
78kingabzpro/wav2vec2-large-xls-r-300m-Urdu-~$0.0034 / minute315.5Mcoming soon
79theainerd/Wav2Vec2-large-xlsr-hindi-~$0.0034 / minute315.5Mcoming soon