Modelos de IA con visión: LLM multimodales para la comprensión de imágenes
Clasificaciones de modelos actualizadas según datos de uso reales.
Los modelos se clasifican según el total de tókenes de prompts y respuestas procesados a través de la API de OpenRouter durante los últimos 7 días. Se muestran los 10 modelos mejor clasificados. Las clasificaciones miden el uso en OpenRouter y reflejan la adopción, no la calidad del modelo ni su rendimiento en pruebas de evaluación.
Los modelos de visión son LLM multimodales que analizan imágenes, leen documentos, interpretan gráficos y responden preguntas sobre contenido visual junto con texto. Esta colección clasifica los modelos con capacidades de visión según su uso en OpenRouter durante la última semana. Los mejores modelos actuales son DeepSeek V4.1 Flash 1017, Hy4 y DeepSeek V4.1 Flash. Acceda a modelos de Anthropic, Google, OpenAI y otros proveedores mediante una única API y compare la longitud del contexto, los precios y las capacidades.
Los mejores modelos de visión en OpenRouter

DeepSeek: DeepSeek V4.1 Flash 1017
DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the deepseek-flash API model name. V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on output from a 552B-parameter backbone, an asymmetric split that keeps per-token compute low relative to the model's total size. Image understanding is native to the architecture, with visual and text embeddings trained jointly from the start of pre-training rather than added afterward as in the earlier experimental V4 Flash Vision Exp.
It is suited for coding, terminal, and computer-use agents, along with long-horizon tasks that must run to completion across many steps and long-context analysis. Compressed KV caching cuts cache memory to roughly a quarter of the previous Flash generation, significantly reducing costs on agentic workloads. DeepSeek positions it as the cost-efficient tier of the V4.1 family and reports that it exceeds V4 Pro on performance, speed, and task completion time.

Tencent: Hy4
The official release of Hy4 delivers comprehensive and substantial improvements over Hy4 Preview, with particularly strong gains in agentic capabilities and performance across a wide range of long-horizon, real-world productivity tasks. It also introduces visual understanding for the first time.

DeepSeek: DeepSeek V4.1 Flash
DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on output from a 552B-parameter backbone, an asymmetric split that keeps per-token compute low relative to the model's total size. Image understanding is native to the architecture, with visual and text embeddings trained jointly from the start of pre-training rather than added afterward as in the earlier experimental V4 Flash Vision Exp.
It is suited for coding, terminal, and computer-use agents, along with long-horizon tasks that must run to completion across many steps and long-context analysis. Compressed KV caching cuts cache memory to roughly a quarter of the previous Flash generation, significantly reducing costs on agentic workloads. DeepSeek positions it as the cost-efficient tier of the V4.1 family and reports that it exceeds V4 Pro on performance, speed, and task completion time.
Z.ai: GLM 5.3 Flash
GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks. Its hybrid sparse and linear attention architecture maintains accurate long-context behavior while reducing compute overhead.
OpenAI: GPT-6 Luna
GPT-6 Luna is the fast, cost-efficient model in OpenAI's GPT-6 series, positioned below GPT-6 Sol. It is suited for high-volume and latency-sensitive workloads such as chat, classification, and lightweight agentic tasks, and at higher reasoning effort it can take on complex software engineering and computer-use tasks that previously called for a Sol-tier model. It shares the GPT-6 family's gains in factual reliability and its clearer, more concise communication style.
Anthropic: Claude Opus 5.5
Claude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, succeeding Claude Opus 5. It is particularly strong at multi-step changes in large codebases, code review and bug finding, financial and scientific analysis, and reading dense charts, diagrams, and screenshots, and it is more careful than its predecessor about only stating figures and citing sources it can back up.
The model completes comparable tasks in fewer steps and with fewer tokens than Opus 5, and reports on its work in plainer language, with clear updates on what it did, what it found, and what it needs from the user. Thinking is always adaptive, so effort is the main lever for trading off depth, latency, and cost, and lower effort settings remain effective for latency-sensitive workloads.
StepFun: Step 5 Preview
Step 5 Preview is StepFun's flagship model for agentic work, built on a sparse Mixture-of-Experts architecture (27B active / 600B total parameters). It performs strongly in software engineering and professional knowledge work, with particular strength in finance.
It is designed for extended tasks that span large codebases and documents, using tools and refining results over multiple steps.
Xiaomi: MiMo-V2.6-Flash
MiMo-V2.6-Flash is an open-source foundation model developed by Xiaomi. Built on a Mixture-of-Experts architecture with 309B total parameters and 15B activated per token, it employs a hybrid attention mechanism for greater computational efficiency. The model features a 1M-token context window and native multimodal capabilities. Optimized for agentic workflows, it delivers strong performance across coding, visual, general, and research scenarios, excelling at complex, long-horizon tasks with robust generalization across a diverse range of agent harnesses.
OpenAI: GPT-6.1 Sol
GPT-6.1 Sol is an upgrade to GPT-6 Sol from OpenAI, positioned below the flagship GPT-6 Astra in the GPT-6 series. It is suited for agentic coding, computer use, document-heavy professional work, and multi-step business workflow automation, and approaches Astra-level results on these tasks at a much lower cost. Compared with GPT-6 Sol, it makes fewer factual errors and is more reliable at respecting explicit restrictions and user intent during agentic tasks.

MoonshotAI: Kimi K3.1 Pro
Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. It is built for long-horizon coding, 3D creation, financial research, and end-to-end agentic knowledge work, with native image and video understanding.
Preguntas frecuentes
Añada una parte de contenido input_image cuyo image_url sea una URL HTTP(S) o una URL de datos en base64 a una solicitud POST /api/v1/responses, o una parte de contenido image_url a POST /api/v1/chat/completions. Coloque primero el prompt de texto y después las imágenes. La guía de entradas de imagen muestra ambos formatos de solicitud.
Sí. Añada cada imagen como una parte de contenido independiente en el mismo mensaje de usuario. El número de imágenes que acepta una solicitud varía según el proveedor y el modelo, y los mensajes del sistema deben contener solo texto.
Sí. Envíe un PDF como entrada file_url o file_data en base64 en POST /api/v1/responses, o como parte de contenido de archivo en Chat Completions; la guía de entradas PDF explica los motores de análisis y su coste.
Sí. Una clave de API de OpenRouter funciona con todos los modelos de visión de esta página y con todos los demás modelos de OpenRouter. Cree una clave en su página de claves de API.