23 comments

  • homarp 17 hours ago

    Running Voxtral-Mini-3B-2507 on GPU requires ~9.5 GB of GPU RAM in bf16 or fp16.

    Running Voxtral-Small-24B-2507 on GPU requires ~55 GB of GPU RAM in bf16 or fp16.

  • kamranjon 8 hours ago

    Im pretty excited to play around with this. I’ve worked with whisper quite a bit, it’s awesome to have another model in the same class and from Mistral, who tend to be very open. I’m sure unsloth is already working on some GGUF quants - will probably spin it up tomorrow and try it on some audio.

  • ipsum2 13 hours ago

    24B is crazy expensive for speech transcription. Conspicuously no comparison with Parakeet, a 600M param model thats currently dominating leaderboards (but only for English)

    • azinman2 7 hours ago

      But it also includes world knowledge, can do tool calls, etc. It’s an omnimodel

  • sheerun 9 hours ago

    In demo they mention polish prononcuation is pretty bad, spoken as if second language of english-native speaker. I wonder if it's the same for other languages. On the other hand whispering-english is hillariously good, especially different emotions.

    • Raed667 6 hours ago

      It is insane how good the "French man speaking English" demo is. It captures a lot of subtleties

  • danelski 20 hours ago

    They claim to undercut competitors of similar quality by half for both models, yet they released both as Apache 2.0 instead of following smaller - open, larger - closed strategy used for their last releases. What's different here?

    • halJordan 13 hours ago

      They didn't release voxtral large so your question doesn't really make sense

      • danelski 6 hours ago

        It's about what their top offering is at the moment, not having Large in name. Mistral Medium 3 is notably not Mistral Large 3, but it was released as API-only.

    • wmf 14 hours ago

      They're working on a bunch of features so maybe those will be closed. I guess they're feeling generous on the base model.

    • Havoc 14 hours ago

      Probably not looking to directly compete in transcription space

  • GaggiX 18 hours ago

    There is also a Voxtral Small 24B small model available to be downloaded: https://huggingface.co/mistralai/Voxtral-Small-24B-2507

  • lostmsu 16 hours ago

    Does it support realtime transcription? What is the ~latency?

    • rolisz an hour ago

      Unlikely. The small model is much larger than whisper (which is already hard to use for realtime)

  • homarp 19 hours ago
    • homarp 19 hours ago

      Running Voxtral-Mini-3B-2507 on GPU requires ~9.5 GB of GPU RAM in bf16 or fp16.

      Running Voxtral-Small-24B-2507 on GPU requires ~55 GB of GPU RAM in bf16 or fp16.

  • lostmsu 16 hours ago

    My Whisper v3 Large Turbo is $0.001/min, so their price comparison is not exactly perfect.