{"id":"voxtral-small","nom":"Voxtral Small","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-Small-24B-2507","page":"https://huggingface.co/mistralai/Voxtral-Small-24B-2507","essayer":"https://docs.mistral.ai/studio/audio/speech_to_text"},"fiche_auto":false,"page":"https://quelleia.com/voix/modeles/voxtral-small/","releve":"2026-10-19","en_bref":"9e sur 39 en transcription en français (ex æquo à valeur égale) ; 9e sur 51 en transcription en anglais (Artificial Analysis) ; 29e sur 53 en transcription en anglais (Open ASR) (ex æquo à valeur égale).","presentation":"Modèle audio de Mistral AI bâti sur Mistral Small 3, avec des poids publiés, qui transcrit, traduit et comprend la parole. Il répond à des questions sur un enregistrement, le résume et peut déclencher des fonctions à partir d'une consigne orale.","sortie":"2025-07-15","taille":"24 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-Small-24B-2507/raw/main/README.md (« Voxtral Small is an enhancement of [Mistral Small 3](https://huggingface.co/mistralai/Mistral-Small-24B-Base-2501), incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance. »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://huggingface.co/mistralai/Voxtral-Small-24B-2507/raw/main/README.md (« - **Built-in Q&A and summarization**: Supports asking questions directly through audio. »), 17/10/2026"},{"appuie":["presentation"],"lu":"https://huggingface.co/mistralai/Voxtral-Small-24B-2507/raw/main/README.md (« - **Function-calling straight from voice**: Enables direct triggering of backend functions, workflows, or API calls based on spoken user intents »), 17/10/2026"},{"appuie":["taille"],"lu":"https://huggingface.co/mistralai/Voxtral-Small-24B-2507/raw/main/README.md (« # Voxtral Small 1.0 (24B) - 2507 »), 17/10/2026"},{"appuie":["langues"],"lu":"https://huggingface.co/mistralai/Voxtral-Small-24B-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-small-25-07 (« July 15, 2025 Blog Technical Report GA Apache 2.0 v 25.07 Voxtral Small Our first model with audio input capabilities for instruct use cases. »), 17/10/2026"}],"classements":[{"id":"transcription/francais","nom":"Transcription en français","rang":9,"sur":39,"valeur":4.11,"unite":"%","chiffres":{"rtfx":89.24,"jeux":{"Common Voice":5.5,"FLEURS":3.61,"MLS":3.22}},"voisins":["resonant-1-flash","modulate-multilingual"],"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/artificial-analysis","nom":"Transcription en anglais, indice AA-WER","rang":9,"sur":51,"valeur":2.77,"unite":"%","chiffres":{"prix_pour_1000_minutes":4,"jeux":{"AA-AgentTalk":2.27,"Earnings22":4.78,"VoxPopuli":1.75},"date_sortie":"2025-07-15","hebergeur":"Mistral"},"voisins":["zoom-scribe-v2-pro","google-gemini-3-1-pro-preview-high"],"source":{"nom":"Artificial Analysis","mention":"Source : Artificial Analysis","url":"https://artificialanalysis.ai/speech-to-text/models/mistral","releve":"2026-10-19"}},{"id":"transcription/open-asr-anglais","nom":"Transcription en anglais, Open ASR","rang":29,"sur":53,"valeur":6.38,"unite":"%","chiffres":{"rtfx":101.06,"jeux":{"AMI-Cleaned":13.19,"Earnings22-Cleaned-AA-chunked":7.07,"Gigaspeech-Cleaned":8.33,"LS Clean":1.23,"LS Other":2.8,"Private (conversational)":14.7,"Private (scripted)":3.22,"SPGISpeech":1.84,"URGENT24":12.35,"Voice Arena Monsoon":3.49,"Voxpopuli-AA-Cleaned":2}},"voisins":["ibm-granite-speech-5-0-470m-turboctc","ibm-granite-4-0-1b-speech"],"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-Small-24B-2507","artificial_analysis":"Voxtral Small"}}