Idéal pour : (Vision + Text)
Ranked on vision benchmark accuracy, context window, and combined per-query cost for image + text workloads.
Mis à jour May 2026. Top 3 ce mois-ci : Qwen3.5 Plus 2026-02-15, Qwen3.5 397B A17B, GPT-4o (2024-11-20).
Podium
Notre classement
Multimodal workloads — document parsing, screenshot understanding, chart reading — demand a model that can handle dense visual information without hallucinating fields that are not present. We weight MMMU for vision reasoning and DocVQA for document tasks, alongside price because vision tokens are priced differently per provider.
Our full methodology is published on the page méthodologie.
Piliers et poids :
Full ranking
| Rang | Modèle | Fournisseur | Entrée $/1M | Sortie $/1M | Contexte |
|---|---|---|---|---|---|
| 1 | Qwen3.5 Plus 2026-02-15 | Qwen | $0.26 | $1.56 | 1,000,000 |
| 2 | Qwen3.5 397B A17B | Qwen | $0.39 | $2.34 | 262,144 |
| 3 | GPT-4o (2024-11-20) | OpenAI | $2.50 | $10.00 | 128,000 |
| 4 | MiniMax-01 | MiniMax | $0.20 | $1.10 | 1,000,192 |
| 5 | Claude Sonnet 4.5 | Anthropic | $3.00 | $15.00 | 1,000,000 |
| 6 | Qwen3.5-122B-A10B | Qwen | $0.26 | $2.08 | 262,144 |
| 7 | Qwen3.5-27B | Qwen | $0.20 | $1.56 | 262,144 |
| 8 | Llama 4 Maverick | Meta | $0.15 | $0.60 | 1,048,576 |
| 9 | Gemma 4 31B | $0.00 | $0.00 | 262,144 | |
| 10 | Gemma 4 31B | $0.13 | $0.38 | 262,144 |
Field notes
Verify the provider's image-size and token-per-image policy before committing — pricing varies dramatically.
For structured extraction, combine vision with JSON mode.
Chart/graph understanding is still fragile. Add a text-only sanity check if the downstream consequence is high.
FAQ
The questions teams ask before picking a model for (vision + text).
Get instant answers from our AI agent
