Modelos de IA gratuitos na OpenRouter
Classificações de modelos atualizadas com base em dados reais de uso.
Os modelos são classificados pelo total de tokens de prompt e de resposta processados pela API do OpenRouter nos últimos 7 dias. Exibindo os 16 melhores modelos. As classificações medem o uso no OpenRouter e refletem a adoção, não a qualidade do modelo ou o desempenho em testes de referência.
Os modelos de IA gratuitos mais usados estão classificados abaixo pelo uso real no OpenRouter na última semana, e todos eles têm preço de US$ 0 para prompts. Os melhores modelos atualmente são Nemotron 3 Ultra (free), Nemotron 3.5 Super VL (free) e ERNIE-5.5-Flash-Preview (free). Os modelos gratuitos permitem criar protótipos, aprender e executar projetos paralelos com um orçamento quase nulo, e o OpenRouter continua ampliando a capacidade gratuita ao integrar provedores e cobrir os custos diretamente. Se você preferir não escolher, openrouter/free encaminha cada solicitação para um modelo gratuito adequado a ela.
Principais modelos gratuitos no OpenRouter
NVIDIA: Nemotron 3 Ultra (free)
NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it supports text input and output with a context window of up to 1M tokens. It is suited for long-running agentic workflows, including agent orchestration, coding agents, deep research, and complex enterprise tasks.
It is particularly strong at multi-step reasoning and planning, with high-throughput inference designed for high-volume agent pipelines. It is part of the NVIDIA Nemotron family of open models for agentic AI.
NVIDIA: Nemotron 3.5 Super VL (free)
NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total. It takes text, images, documents, GUI screenshots, and video as input and returns text, and is built for long-running agents that combine visual understanding with reasoning and tool use.
It uses a hybrid Mamba-Transformer architecture with multi-token prediction and supports a 1M-token context. Typical uses include document intelligence, web research, coding validation, customer service, and video understanding. Reasoning is on by default and can be turned off or given a token budget.
Baidu: ERNIE-5.5-Flash-Preview (free)
ERNIE 5.5 Flash Preview is the latest foundation model in the ERNIE family, with comprehensive improvements in coding, agentic capabilities, knowledge work, and deep research. It has 280 billion total parameters, activates 14 billion parameters per token, and supports a context window of 1 million tokens. Its new architecture combines strong model capabilities with efficient inference, delivering generation speeds of more than 200 tokens per second.
NVIDIA: Nemotron 3.5 Lightning (free)
NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, with 3B active parameters out of 30B total. It is suited for high-throughput agentic workloads and specialized tasks that benefit from domain-specific customization.
Thinking Machines: Inkling 2 Small (free)
Inkling 2 Small is an open-weight multimodal mixture-of-experts reasoning model from Thinking Machines Lab, with 17B active parameters out of 594B total. It is the first model in the second Inkling generation. It was trained with reinforcement learning on long-horizon coding tasks across many agent harnesses, and is suited for agentic coding, autonomous research, agentic knowledge work and retrieval-augmented generation, with English and general multilingual support. It accepts text and image input (no audio), supports tool calling and variable reasoning effort, and has a 1M-token context window.

Dots Studio: Dots3-Note Preview (free)
Dots3-Note Preview is an open-weight mixture-of-experts model from Dots Studio, with 16B active parameters out of 280B total. It is the lightest model in the Dots 3 family and is suited for reasoning, coding, multimodal understanding, long-context processing, and multi-step agent workflows.
Particle AI: Particle-1 (free)
Particle-1 is a token-efficient agentic coding model from Particle AI, post-trained on DeepSeek V4.1 Flash. It is suited for software engineering work such as terminal workflows, codebase changes, and data engineering tasks, and is trained to complete them with substantially fewer reasoning and tool-call tokens while keeping accuracy close to its base model.

Cohere: North Mini Code (free)
North Mini Code is Cohere's first agentic coding model and the debut of its North family. A sparse mixture-of-experts model with 30B total parameters and 3B active, it is optimized for code generation, agentic software engineering, and terminal tasks, and is trained to generalize across agent harnesses such as OpenCode and SWE-Agent.
It offers a 256K-token context window with up to 64K tokens of output, supports interleaved reasoning and tool use via JSON schema, and is released open-weight under the Apache 2.0 license. Its small active-parameter footprint enables low-latency inference, including on local hardware.
Recursive: Flint (free)
Flint is an open-weight mixture-of-experts model from Recursive, with about 2.5B active parameters out of 40B total, built by Recursive's autoresearch system Eureka Machine for coding agents. It is suited for investigating codebases, fixing bugs, implementing features, and writing tests.
inclusionAI: Ling 3.2 Flash (free)
Ling 3.2 Flash is a hybrid reasoning model from inclusionAI that accepts text and image input, with a 1M-token context window and up to 200 images per request. Thinking can be switched off per request.

Perplexity: Decider V1.3 (free)
Decider V1.3 is a decision model from Perplexity, succeeding Decider V1.2. Instead of generating text, it reads content passed as state (text, JSON, or images) and returns typed, probabilistic answers to one or more named questions in a single request: the probability of yes for a yes/no question (noul), a probability for every option plus the most likely one (choice, up to 255 options), or a probability for every level of an ordered rubric plus the expected score (score).
It is built for classification, routing, moderation, and rubric grading where application code thresholds the returned numbers rather than parsing a chat reply.
V1.3 also adds locate, served on the Extractions route: it pulls text and list fields out of text state and returns each value with its source span and a confidence. For a smaller, lower-cost option with the same contract, see Decider Flash V1.3.
NVIDIA: Nemotron 3 Nano Omni (free)
NVIDIA Nemotron™ 3 Nano Omni is a 30B-A3B open multimodal model designed to function as a perception and context sub-agent in enterprise agent systems. It accepts text, image, video, and audio inputs and produces text output, enabling agents to perceive and reason across modalities in a single inference loop.
Built on a hybrid MoE Transformer-Mamba architecture with Conv3D video layers and Efficient Video Sampling (EVS), it delivers approximately 2× higher throughput and 2.5× lower compute for video reasoning versus separate vision + speech pipelines. It supports up to 300K context length and a 16,384 reasoning budget, with extended thinking enabled via reasoning.enabled on OpenRouter.

LiquidAI: LFM3-1.7B (free)
A compact 1.7B-parameter text model with a 256K-token context, post-trained for agents, tool calling, and structured outputs. It reasons before it answers. Prompts and outputs may be retained and used to train Liquid models.

Perplexity: Decider Flash V1.3 (free)
Decider Flash V1.3 is a decision model from Perplexity, a smaller, lower-cost sibling of Decider V1.3 with the same API contract. Instead of generating text, it reads content passed as state (text, JSON, or images) and returns typed, probabilistic answers to one or more named questions in a single request: the probability of yes for a yes/no question (noul), a probability for every option plus the most likely one (choice, up to 255 options), or a probability for every level of an ordered rubric plus the expected score (score).
It suits high-volume classification, routing, and moderation where application code thresholds the returned numbers rather than parsing a chat reply.
It also adds locate, served on the Extractions route, pulling text and list fields out of text state with source spans and confidences.
Respan: Span-01 Lite
Span-01 Lite is the free, lighter tier of Span-01, a behavior scoring model from Respan. It returns, for each plain-language behavior you define, the probability that the behavior is present in a conversation span, and is suited for high-volume evaluation and monitoring where cost matters more than peak accuracy.
NVIDIA: Llama Nemotron Rerank VL 1B V2 (free)
Llama Nemotron Rerank VL 1B V2 is a 1.7B multimodal reranking model from NVIDIA. It evaluates the relevance of document images and text against user queries, designed for vision RAG pipelines handling charts, tables, infographics, and mixed-media documents. Functions as a cross-encoder that accepts text queries paired with image, text, or combined document inputs, delivering approximately 6-7% recall improvements over embedding-only baselines on visual document retrieval benchmarks.
Perguntas frequentes
Sim. Todos os modelos nesta página têm preço de $0, portanto seu uso não é cobrado. As variantes gratuitas, cujos IDs terminam em :free, não exigem a compra de créditos, enquanto outros modelos de $0 ainda podem exigir uma conta com saldo. O uso gratuito está sujeito a limites de taxa.
As variantes gratuitas, cujos IDs terminam em :free, têm um limite de 20 solicitações por minuto e 50 solicitações por dia, que aumenta para 1.000 solicitações por dia quando sua conta tiver comprado US$ 10 em créditos (o limite maior começa em US$ 9 para absorver as taxas de recarga). Tentativas malsucedidas contam para o limite diário. Outros modelos de US$ 0 não estão sujeitos a esses limites, mas ainda podem atingir limites por modelo definidos pelo OpenRouter ou pelo provedor. Consulte a documentação de limites de taxa.
Para modelos de chat, envie uma requisição ao mesmo endpoint de Chat Completions compatível com OpenAI que você usa para modelos pagos, com o ID do modelo gratuito no campo model. Outros tipos de modelos usam suas próprias APIs, como /api/v1/rerank para modelos de reclassificação e /api/alpha/decisions para modelos de decisão, portanto consulte a página do modelo para encontrar o endpoint correto. As variantes gratuitas de modelos pagos terminam em :free.
Sim. Defina o modelo como openrouter/free e cada solicitação será roteada para um modelo gratuito adequado a ela, para que você não precise acompanhar quais modelos gratuitos estão disponíveis no momento.
Não. Uma única chave de API do OpenRouter funciona tanto para modelos gratuitos quanto pagos — crie uma chave na sua página de chaves de API e reutilize-a em todos os modelos.