{"id":"voxtral-mini","nom":"Voxtral Mini","editeur":"Mistral AI","pays":"France","poids_ouverts":true,"poids_selon":null,"licence":"Apache 2.0","licence_selon":null,"liens":{"poids":"https://huggingface.co/mistralai/Voxtral-Mini-3B-2507","page":"https://huggingface.co/mistralai/Voxtral-Mini-3B-2507"},"fiche_auto":false,"page":"https://quelleia.com/voix/modeles/voxtral-mini/","releve":"2026-10-19","en_bref":"22e sur 39 en transcription en français ; 38e sur 53 en transcription en anglais (Open ASR).","presentation":"Petit modèle audio de Mistral AI, bâti sur Ministral 3B, qui transcrit, traduit et comprend la parole, avec des poids publiés. Il traite jusqu'à 30 minutes d'audio en transcription, peut répondre à des questions sur un enregistrement et déclencher des fonctions à la voix.","sortie":"2025-07-15","taille":"3 milliards de paramètres","langues":{"annoncees":{"annoncees":"8 langues, dont le français","nombre":8,"francais":true,"liste":["en","es","fr","pt","hi","de","nl","it"]},"prises_en_charge_selon_open_tts":null,"evaluees":["fr","en"]},"preuves":[{"appuie":["sortie"],"lu":"https://mistral.ai/news/voxtral (« Voxtral July 15, 2025 By Mistral »), 17/10/2026"},{"appuie":["taille"],"lu":"https://mistral.ai/news/voxtral (« These state‑of‑the‑art speech understanding models are available in two sizes-a 24B variant for production-scale applications and a 3B variant for local and edge deployments. Both versions are released under the Apache 2.0 license »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://huggingface.co/mistralai/Voxtral-Mini-3B-2507/raw/main/README.md (« Voxtral Mini is an enhancement of [Ministral 3B](https://mistral.ai/news/ministraux), incorporating state-of-the-art audio input capabilities »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://huggingface.co/mistralai/Voxtral-Mini-3B-2507/raw/main/README.md (« It excels at speech transcription, translation and audio understanding. »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://huggingface.co/mistralai/Voxtral-Mini-3B-2507/raw/main/README.md (« With a 32k token context length, Voxtral handles audios up to 30 minutes for transcription, or 40 minutes for understanding »), 17/10/2026"},{"appuie":["taille"],"lu":"https://huggingface.co/mistralai/Voxtral-Mini-3B-2507/raw/main/README.md (« # Voxtral Mini 1.0 (3B) - 2507 »), 17/10/2026"},{"appuie":["langues"],"lu":"https://huggingface.co/mistralai/Voxtral-Mini-3B-2507/raw/main/README.md (« Automatic language detection and state-of-the-art performance in the world’s most widely used languages (English, Spanish, French, Portuguese, Hindi, German, Dutch, Italian) »), 17/10/2026"},{"appuie":["sortie"],"lu":"https://docs.mistral.ai/models/voxtral-mini-25-07 (« July 15, 2025 Blog Technical Report Deprecated Apache 2.0 v 25.07 Voxtral Mini A mini version of our first audio input model. »), 17/10/2026"}],"classements":[{"id":"transcription/francais","nom":"Transcription en français","rang":22,"sur":39,"valeur":5.96,"unite":"%","chiffres":{"rtfx":158.29,"jeux":{"Common Voice":7.8,"FLEURS":4.13,"MLS":5.96}},"voisins":["alibaba-qwen3-asr-1-7b-hf","soniox-v5-async"],"source":{"nom":"Open ASR Leaderboard","mention":"Source : Open ASR Leaderboard, Hugging Face","url":"https://huggingface.co/spaces/hf-audio/open_asr_leaderboard","releve":"2026-10-19"}},{"id":"transcription/open-asr-anglais","nom":"Transcription en anglais, Open ASR","rang":38,"sur":53,"valeur":7.25,"unite":"%","chiffres":{"rtfx":180.05,"jeux":{"AMI-Cleaned":13.57,"Earnings22-Cleaned-AA-chunked":7.66,"Gigaspeech-Cleaned":8.7,"LS Clean":1.48,"LS Other":3.63,"Private (conversational)":16,"Private (scripted)":3.73,"SPGISpeech":2.15,"URGENT24":15.66,"Voice Arena Monsoon":4.9,"Voxpopuli-AA-Cleaned":2.22}},"voisins":["funaudiollm-fun-asr-nano-2512-hf","nvidia-canary-1b-v2"],"source":{"nom":"Open ASR Leaderboard","mention":"Source : Open ASR Leaderboard, Hugging Face","url":"https://huggingface.co/spaces/hf-audio/open_asr_leaderboard","releve":"2026-10-19"}}],"noms_dans_les_sources":{"hf_audio":"mistralai/Voxtral-Mini-3B-2507"}}