Capacité · 2026-08-14
Modèles d'IA avec entrée visuelle
Modèles acceptant des images en plus du texte — compréhension multimodale.
Qu'est-ce que c'est ?
- Les modèles vision-langage acceptent des images en plus de (ou à la place du) texte.
- La plupart répondent en texte — ce sont des LLMs multimodaux, pas des générateurs d'images.
Pourquoi c'est important
- Cas d'usage : compréhension de documents (scans, PDF, captures d'écran), revue UI/code depuis des captures, Q&A sur photos produit, accessibilité (alt text), imagerie médicale/satellite.
- La facturation inclut souvent un coût par image en plus du coût par token — consultez les tableaux d'offering de chaque fournisseur.
584 modèles avec cette capacité
Top 60 sur 584 affichés. Utilisez le répertoire complet pour filtrer davantage.
Frequently asked questions
How many AI models support entrée d'image?
584 canonical models in our database currently support entrée d'image. The list is regenerated on every data refresh, so it always reflects the latest releases tracked in our catalogue.
What is the cheapest model with entrée d'image?
PaddleOCR-VL from novita-ai is currently the lowest-priced option, at $0.020 per 1M input tokens and $0.020 per 1M output tokens. The full table above is sorted price-ascending.
Which model with entrée d'image has the largest context window?
Llama 4 Scout 17B Instruct (US) (Meta) leads on context at 3.50M tokens. This may matter if you also need long-document understanding alongside entrée d'image.
Which models are available on the most providers?
Production-readiness usually correlates with how many independent providers host the same weights. The top three by provider count are: Kimi K2.6 (63), Kimi K2.5 (50), Kimi K3 (50).
How is entrée d'image different from a regular LLM?
Vision-language models accept image input alongside text. They are multimodal LLMs, not image generators — most reply in text after looking at the image.
How often is this list updated?
Daily. Our data pipeline syncs once a day, regenerates the canonical model list, and rebuilds these pages so newly released models appear within 24 hours.
Explore more
Top models with this capability
- PaddleOCR-VL$0.02 in / $0.02 out
- DeepSeek OCR 2$0.03 in / $0.03 out
- Llama 3.2 11B Vision Instruct$0.06 in / $0.06 out
- Gemma 3 4B IT$0.04 in / $0.08 out
- Gemma 4 E4B IT$0.02 in / $0.10 out
Other capabilities
Best-of lists you might also want
Pricing comparisons
Dernière mise à jour :
Prices in USD per 1M tokens. Unknown means the provider does not publish per-token pricing.
Pricing and capabilities are refreshed daily and reconciled against each provider's official documentation. Always verify critical production decisions with the provider directly.