{"model_id": "microsoft/Phi-4-multimodal-instruct", "name": "Phi-4-multimodal-instruct", "owner": "microsoft", "source_url": "https://huggingface.co/microsoft/Phi-4-multimodal-instruct", "task": "automatic-speech-recognition", "languages": ["multilingual", "ar", "zh", "cs", "da", "nl", "en", "fi", "fr", "de", "he", "hu", "it", "ja", "ko", "no", "pl", "pt", "ru", "es", "sv", "th", "tr", "uk"], "tags": ["transformers", "safetensors", "phi4mm", "text-generation", "nlp", "code", "audio", "automatic-speech-recognition", "speech-summarization", "speech-translation", "visual-question-answering", "phi-4-multimodal", "phi", "phi-4-mini", "custom_code", "multilingual", "ar", "zh", "cs", "da", "nl", "en", "fi", "fr", "de", "he", "hu", "it", "ja", "ko", "no", "pl", "pt", "ru", "es", "sv", "th", "tr", "uk", "arxiv:2503.01743", "arxiv:2407.13833", "license:mit", "eval-results", "region:us"], "license_id": "mit", "license_status": "ouverte_permissive", "license_label": "MIT License", "commercial_use": "possible_a_verifier", "gated": false, "private": false, "downloads": 195684, "likes": 1616, "updated_at": "2025-12-10T20:18:10.000Z", "description": "", "files": [{"name": ".gitattributes", "size": 1612, "format": "autre", "precision": "inconnue"}, {"name": "CODE_OF_CONDUCT.md", "size": 444, "format": "autre", "precision": "inconnue"}, {"name": "LICENSE", "size": 1141, "format": "autre", "precision": "inconnue"}, {"name": "README.md", "size": 65395, "format": "autre", "precision": "inconnue"}, {"name": "SECURITY.md", "size": 2656, "format": "autre", "precision": "inconnue"}, {"name": "SUPPORT.md", "size": 1244, "format": "autre", "precision": "inconnue"}, {"name": "added_tokens.json", "size": 249, "format": "autre", "precision": "inconnue"}, {"name": "config.json", "size": 4631, "format": "autre", "precision": "inconnue"}, {"name": "configuration_phi4mm.py", "size": 11014, "format": "autre", "precision": "inconnue"}, {"name": "data_summary_card.md", "size": 4619, "format": "autre", "precision": "inconnue"}, {"name": "examples/what_is_shown_in_this_image.wav", "size": 112844, "format": "autre", "precision": "inconnue"}, {"name": "examples/what_is_the_traffic_sign_in_the_image.wav", "size": 741454, "format": "autre", "precision": "inconnue"}, {"name": "figures/audio_understand.png", "size": 42645, "format": "autre", "precision": "inconnue"}, {"name": "figures/multi_image.png", "size": 191950, "format": "autre", "precision": "inconnue"}, {"name": "figures/speech_qa.png", "size": 46754, "format": "autre", "precision": "inconnue"}, {"name": "figures/speech_recog_by_lang.png", "size": 90697, "format": "autre", "precision": "inconnue"}, {"name": "figures/speech_recognition.png", "size": 63547, "format": "autre", "precision": "inconnue"}, {"name": "figures/speech_summarization.png", "size": 40951, "format": "autre", "precision": "inconnue"}, {"name": "figures/speech_translate.png", "size": 47677, "format": "autre", "precision": "inconnue"}, {"name": "figures/speech_translate_2.png", "size": 46342, "format": "autre", "precision": "inconnue"}, {"name": "figures/vision_radar.png", "size": 173821, "format": "autre", "precision": "inconnue"}, {"name": "generation_config.json", "size": 190, "format": "autre", "precision": "inconnue"}, {"name": "merges.txt", "size": 2418348, "format": "autre", "precision": "inconnue"}, {"name": "model-00001-of-00003.safetensors", "size": 4997504848, "format": "safetensors", "precision": "inconnue"}, {"name": "model-00002-of-00003.safetensors", "size": 4952333128, "format": "safetensors", "precision": "inconnue"}, {"name": "model-00003-of-00003.safetensors", "size": 1199389232, "format": "safetensors", "precision": "inconnue"}, {"name": "model.safetensors.index.json", "size": 239890, "format": "autre", "precision": "inconnue"}, {"name": "modeling_phi4mm.py", "size": 116057, "format": "autre", "precision": "inconnue"}, {"name": "phi_4_mm.tech_report.02252025.pdf", "size": 5295165, "format": "autre", "precision": "inconnue"}, {"name": "preprocessor_config.json", "size": 482, "format": "autre", "precision": "inconnue"}, {"name": "processing_phi4mm.py", "size": 32775, "format": "autre", "precision": "inconnue"}, {"name": "processor_config.json", "size": 121, "format": "autre", "precision": "inconnue"}, {"name": "sample_finetune_speech.py", "size": 16669, "format": "autre", "precision": "inconnue"}, {"name": "sample_finetune_vision.py", "size": 19617, "format": "autre", "precision": "inconnue"}, {"name": "sample_inference_phi4mm.py", "size": 10498, "format": "autre", "precision": "inconnue"}, {"name": "special_tokens_map.json", "size": 473, "format": "autre", "precision": "inconnue"}, {"name": "speech-lora/adapter_config.json", "size": 465, "format": "autre", "precision": "inconnue"}, {"name": "speech-lora/adapter_model.safetensors", "size": 922782296, "format": "safetensors", "precision": "inconnue"}, {"name": "speech-lora/added_tokens.json", "size": 249, "format": "autre", "precision": "inconnue"}, {"name": "speech-lora/special_tokens_map.json", "size": 473, "format": "autre", "precision": "inconnue"}, {"name": "speech-lora/tokenizer.json", "size": 15524095, "format": "autre", "precision": "inconnue"}, {"name": "speech-lora/tokenizer_config.json", "size": 3248, "format": "autre", "precision": "inconnue"}, {"name": "speech-lora/vocab.json", "size": 3910310, "format": "autre", "precision": "inconnue"}, {"name": "speech_conformer_encoder.py", "size": 110521, "format": "autre", "precision": "inconnue"}, {"name": "tokenizer.json", "size": 15524479, "format": "autre", "precision": "inconnue"}, {"name": "tokenizer_config.json", "size": 3248, "format": "autre", "precision": "inconnue"}, {"name": "vision-lora/adapter_config.json", "size": 464, "format": "autre", "precision": "inconnue"}, {"name": "vision-lora/adapter_model.safetensors", "size": 738232904, "format": "safetensors", "precision": "inconnue"}, {"name": "vision-lora/added_tokens.json", "size": 249, "format": "autre", "precision": "inconnue"}, {"name": "vision-lora/special_tokens_map.json", "size": 473, "format": "autre", "precision": "inconnue"}, {"name": "vision-lora/tokenizer.json", "size": 15524095, "format": "autre", "precision": "inconnue"}, {"name": "vision-lora/tokenizer_config.json", "size": 3248, "format": "autre", "precision": "inconnue"}, {"name": "vision-lora/vocab.json", "size": 3910310, "format": "autre", "precision": "inconnue"}, {"name": "vision_siglip_navit.py", "size": 78218, "format": "autre", "precision": "inconnue"}, {"name": "vocab.json", "size": 3910310, "format": "autre", "precision": "inconnue"}], "parameter_count": 5574460384, "library_name": "transformers", "pipeline_tag": "automatic-speech-recognition", "datasets": [], "metrics": [], "github_url": "", "github_metadata": {}, "paper_url": "https://arxiv.org/abs/2503.01743", "paper_doi": "", "paper_metadata": {}, "card_complete": true, "safetensors": true, "provenance": [{"source": "Hugging Face Hub API", "url": "https://huggingface.co/microsoft/Phi-4-multimodal-instruct", "retrieved_at": "2026-10-09T02:22:28.248079+00:00"}], "retrieved_at": "2026-10-09T02:22:28.248079+00:00", "warnings": ["La licence est déclarée par le producteur ; vérifiez toujours son texte intégral et les conditions associées."], "rank_score": 0.0, "rank_reasons": [], "compatibility": {"state": "insuffisante", "label": "Mémoire probablement insuffisante", "summary": "La mémoire indiquée est inférieure au besoin estimé.", "factors": ["Mémoire estimée : 15.5 Go, marge d’exécution de 30% incluse.", "Espace disque minimal indicatif : 11.9 Go.", "Une exécution CPU peut être possible mais sa vitesse ne peut pas être déduite de ces métadonnées."], "estimate": {"known": true, "precision": "float32", "source": "taille réelle déclarée des fichiers de poids", "base_bytes": 12810242408, "estimated_bytes": 16653315130, "overhead": 0.3}}, "quantization": {"available": false, "variants": [], "recommended": null, "lightest": null, "memory_gb": 0.0, "disk_gb": 0.0}, "editorial_quality_score": null, "editorial_quality_blockers": []}