Open-weight transcription models
63 in the catalogue today. Every one with what it costs, who sells it and where it stands.
Models that turn recorded speech into text. They are metered by the minute or the second of audio, so cost follows the length of the recording and not the difficulty of it. What separates them is languages covered, whether they mark who is speaking, and whether they run in real time or only on a finished file — a model that is excellent on a podcast may be unusable on a live call.
Weights published under a licence that lets you run them where you like and sell what you build — Apache 2.0, MIT and their kin impose little beyond keeping the notice. This is the list to start from if you need the model on your own hardware, in your own region, or simply want no third party between you and it. The trade is that hosting is now your problem, and the prices shown beside each one are what somebody else charges to do it for you.
- Breeze-ASR-25MediaTek-Research
- Fun-ASR-Nano-2512QwenAudio
- GLM-ASR-Nano-2512Z.ai
- GigaAM-v3ai-sage
- Hojo-ASR-V1HojoAI7th of 65
- MOSS-Transcribe-DiarizeOpenMOSS-Team17th of 65
- MOSS-Transcribe-preview-2BOpenMOSS-Team19th of 65
- MiMo-V2.5-ASRXiaomi$0.0012–$0.0012per minute1 selling
- Parakeet TDT 0.6B v3NVIDIA22nd of 65
- Qwen3-ASR-0.6BAlibaba$0.000003per second1 selling
- Qwen3-ASR-1.7BAlibaba8th of 65$0.000008per second1 selling
- Qwen3-ForcedAligner-0.6BAlibaba
- VibeVoice-ASRMicrosoft
- VibeVoice-ASR-BitNetMicrosoft
- VibeVoice-ASR-Streaming-7BMicrosoft
- Voxtral Small 24B 2507Mistral AI$0.1→$0.3per Mtok in / out4 selling
- Voxtral-Mini-3B-2507Mistral AI$0.04→$0.04per Mtok in / out2 selling
- Voxtral-Mini-4B-Realtime-2602Mistral AI
- Whisper Large V3 TurboOpenAI$0.000011per second5 selling
- Whisper Large v3OpenAI47th of 65$0.000031per second2 selling
- anime-whisperlitagin
- ast-finetuned-audioset-10-10-0.4593Massachusetts Institute of Technology
- canary-1b-flashNVIDIA23rd of 65
- canary-1b-v2NVIDIA
- canary-qwen-2.5bNVIDIA12th of 65
- clap-htsat-fusedLAION eV
- cohere-transcribe-03-2026Cohere21st of 65$0.000069per second1 selling
- cohere-transcribe-arabic-07-2026Cohere Labs
- distil-large-v2Whisper Distillation
- distil-large-v3Whisper Distillation
- emotion-recognition-wav2vec2-IEMOCAPSpeechBrain
- faster-whisper-large-v2guillaumekln
- faster-whisper-large-v3Systran
- faster-whisper-large-v3-turbo-ct2deepdml
- granite-4.0-1b-speechIBM watsonx.ai25th of 65
- granite-speech-3.3-8bIBM Granite
- granite-speech-4.1-2bIBM watsonx.ai14th of 65
- granite-speech-4.1-2b-narIBM watsonx.ai18th of 65
- higgs-audio-v3-sttBoson AI9th of 65$0.0025per minute1 selling
- indic-conformer-600m-multilingualAI4Bharat
- lang-id-voxlingua107-ecapaSpeechBrain
- parakeet-rnnt-1.1bNVIDIA
- parakeet-tdt-0.6b-v2NVIDIA20th of 65
- speaker-diarizationpyannote
- speaker-diarization-3.0pyannote
- speaker-diarization-3.1pyannote
- speaker-diarization-community-1pyannote
- voice-activity-detectionpyannote
- wav2vec2-base-960hAI at Meta
- wav2vec2-large-960h-lv60-selffacebook
- wav2vec2-large-xlsr-53-englishjonatasgrosman
- wav2vec2-lg-xlsr-en-speech-emotion-recognitionehcalabres
- whisper-baseOpenAI$0.0005per second1 selling
- whisper-base-ar-qurantarteel-ai
- whisper-base.enOpenAI$0.0005per second1 selling
- whisper-largeOpenAI
- whisper-large-v2OpenAI
- whisper-medium.enOpenAI$0.0005per second1 selling
- whisper-smallOpenAI
- whisper-small.enOpenAI$0.0005per second1 selling
- whisper-tinyOpenAI
- whisper-tiny.enOpenAI$0.0005per second1 selling
- whisper.cppggerganov