{"model_id": "sbintuitions/sarashina2.2-vision-3b", "name": "sarashina2.2-vision-3b", "owner": "sbintuitions", "source_url": "https://huggingface.co/sbintuitions/sarashina2.2-vision-3b", "task": "image-to-text", "languages": ["ja", "en"], "tags": ["transformers", "safetensors", "sarashina2_vision", "text-generation", "multimodal", "vision-language", "image-to-text", "custom_code", "ja", "en", "arxiv:2404.07824", "arxiv:2403.19454", "arxiv:2410.17250", "arxiv:2007.00398", "arxiv:2104.12756", "base_model:sbintuitions/sarashina2.2-3b-instruct-v0.1", "base_model:finetune:sbintuitions/sarashina2.2-3b-instruct-v0.1", "license:mit", "region:us"], "license_id": "mit", "license_status": "ouverte_permissive", "license_label": "MIT License", "commercial_use": "possible_a_verifier", "gated": true, "private": false, "downloads": 853, "likes": 20, "updated_at": "2026-08-31T02:17:46.000Z", "description": "", "files": [{"name": ".gitattributes", "size": null, "format": "autre", "precision": "inconnue"}, {"name": "LICENSE", "size": null, "format": "autre", "precision": "inconnue"}, {"name": "README.md", "size": null, "format": "autre", "precision": "inconnue"}, {"name": "chat_template.json", "size": null, "format": "autre", "precision": "inconnue"}, {"name": "config.json", "size": null, "format": "autre", "precision": "inconnue"}, {"name": "configuration_sarashina2_vision.py", "size": null, "format": "autre", "precision": "inconnue"}, {"name": "generation_config.json", "size": null, "format": "autre", "precision": "inconnue"}, {"name": "model.safetensors", "size": null, "format": "safetensors", "precision": "inconnue"}, {"name": "modeling_sarashina2_vision.py", "size": null, "format": "autre", "precision": "inconnue"}, {"name": "preprocessor_config.json", "size": null, "format": "autre", "precision": "inconnue"}, {"name": "processing_sarashina2_vision.py", "size": null, "format": "autre", "precision": "inconnue"}, {"name": "processor_config.json", "size": null, "format": "autre", "precision": "inconnue"}, {"name": "sample.jpg", "size": null, "format": "autre", "precision": "inconnue"}, {"name": "special_tokens_map.json", "size": null, "format": "autre", "precision": "inconnue"}, {"name": "tokenizer.json", "size": null, "format": "autre", "precision": "inconnue"}, {"name": "tokenizer.model", "size": null, "format": "autre", "precision": "inconnue"}, {"name": "tokenizer_config.json", "size": null, "format": "autre", "precision": "inconnue"}], "parameter_count": 3801475696, "library_name": "transformers", "pipeline_tag": "image-to-text", "datasets": [], "metrics": [], "github_url": "", "github_metadata": {}, "paper_url": "https://arxiv.org/abs/2404.07824", "paper_doi": "", "paper_metadata": {}, "card_complete": true, "safetensors": true, "provenance": [{"source": "Hugging Face Hub API", "url": "https://huggingface.co/sbintuitions/sarashina2.2-vision-3b", "retrieved_at": "2026-08-31T13:44:33.617609+00:00"}], "retrieved_at": "2026-08-31T13:44:33.617609+00:00", "warnings": ["La licence est déclarée par le producteur ; vérifiez toujours son texte intégral et les conditions associées.", "Le producteur demande une autorisation ou l’acceptation de conditions avant téléchargement."], "rank_score": 0.0, "rank_reasons": [], "compatibility": {"state": "acces_restreint", "label": "Accès restreint", "summary": "Le modèle est privé ou soumis à une acceptation préalable.", "factors": ["L’accès aux fichiers doit d’abord être autorisé par le producteur."], "estimate": {"known": true, "precision": "float32", "source": "3,801,475,696 paramètres × 4 octet(s)", "base_bytes": 15205902784, "estimated_bytes": 19767673619, "overhead": 0.3}}}