人工智能模型

The simplest way to get free inference. openrouter/free is a router that selects free models at random from the models available on OpenRouter. The router smartly filters for models that...

Step 3.5 Flash is StepFun's most capable open-source foundation model. Built on a sparse Mixture of Experts (MoE) architecture, it selectively activates only 11B of its 196B parameters per token....

Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm. Built on Kimi K2 with continued pretraining over approximately 15T mixed...

Solar Pro 3 is Upstage's powerful Mixture-of-Experts (MoE) language model. With 102B total parameters and 12B active parameters per forward pass, it delivers exceptional performance while maintaining computational efficiency. Optimized...

MiniMax M2-her is a dialogue-first large language model built for immersive roleplay, character-driven chat, and expressive multi-turn conversations. Designed to stay consistent in tone and personality, it supports rich message...

Palmyra X5 is Writer's most advanced model, purpose-built for building and scaling AI agents across the enterprise. It delivers industry-leading speed and efficiency on context windows up to 1 million...

The gpt-audio model is OpenAI's first generally available audio model. 新的快照具有升级的解码器，可实现更自然的声音并保持更好的语音一致性. Audio is priced...

A cost-efficient version of GPT Audio. 新的快照具有升级的解码器，可实现更自然的声音并保持更好的语音一致性. 输入的定价为 $0.60 per million...

As a 30B-class SOTA model, GLM-4.7-Flash 提供了平衡性能和效率的新选项. 它针对代理编码用例进行了进一步优化, strengthening coding capabilities, long-horizon task planning,...

GPT-5.2-Codex 是 GPT-5.1-Codex 的升级版本，针对软件工程和编码工作流程进行了优化. 它专为交互式开发会话和长时间的开发而设计。, independent execution of complex engineering tasks....

种子 1.6 Flash是字节跳动种子公司推出的超快速多模态深度思考模型, supporting both text and visual understanding. It features a 256k context window and can generate outputs of...

种子 1.6 是字节跳动种子团队发布的通用模型. 它结合了多模式功能和自适应深度思维以及 256K 上下文窗口.

MiniMax-M2.1 is a lightweight, 最先进的大型语言模型，针对编码进行了优化, agentic workflows, and modern application development. 仅与 10 billion activated parameters, it delivers a major jump in real-world...

GLM-4.7 is Z.ai’s latest flagship model, featuring upgrades in two key areas: 增强的编程能力和更稳定的多步推理/执行. It demonstrates significant improvements in executing complex agent tasks while...

Gemini 3 Flash Preview is a high speed, 专为代理工作流程设计的高价值思维模型, 多轮聊天, 和编码协助. It delivers near Pro level reasoning and tool...

Gemini 3 Flash Preview is a high speed, 专为代理工作流程设计的高价值思维模型, 多轮聊天, 和编码协助. It delivers near Pro level reasoning and tool...

NVIDIA Nemotron 3 Nano 30B A3B是一种小语言MoE模型，具有最高的计算效率和准确性，可供开发人员构建专门的代理AI系统. The model is fully...

GPT-5.2 聊天 (又名即时) 是最快的, lightweight member of the 5.2 家庭, 针对低延迟聊天进行了优化，同时保留了强大的通用智能. It uses adaptive reasoning to selectively “think” on...

GPT-5.2 Pro is OpenAI’s most advanced model, 与 GPT-5 Pro 相比，在代理编码和长上下文性能方面提供了重大改进. 它针对需要逐步推理的复杂任务进行了优化,...

GPT-5.2 Pro is OpenAI’s most advanced model, 与 GPT-5 Pro 相比，在代理编码和长上下文性能方面提供了重大改进. 它针对需要逐步推理的复杂任务进行了优化,...

GPT-5.2是GPT-5系列中最新的前沿级型号, 与 GPT-5.1 相比，提供更强的代理和长上下文性能. 它使用自适应推理来动态分配计算, responding quickly...

GPT-5.2是GPT-5系列中最新的前沿级型号, 与 GPT-5.1 相比，提供更强的代理和长上下文性能. 它使用自适应推理来动态分配计算, responding quickly...

德夫斯特拉尔 2 是 Mistral AI 专门从事代理编码的最先进的开源模型. 它是一个123B参数的密集变压器模型，支持256K上下文窗口. 德夫斯特拉尔 2 supports exploring...

The relace-search model uses 4-12 “view_file”和“grep”工具并行探索代码库并根据用户请求返回相关文件. 与 RAG 相比, relace-search performs agentic...

GLM-4.6V 是一种大型多模态模型，专为跨图像的高保真视觉理解和长上下文推理而设计, 文件, 和混合媒体. It supports up to 128K tokens, processes complex page layouts...

将您的自然语言请求转换为结构化 OpenRouter API 请求对象. 描述您希望通过 AI 模型实现什么目标, Body Builder 将构建适当的 API 调用. 例子:...

GPT-5.1-Codex-Max是OpenAI最新的代理编码模型, designed for long-running, high-context software development tasks. It is based on an updated version of the 5.1 reasoning stack and trained on agentic...

诺瓦 2 Lite 是一个快速, 适用于可处理文本的日常工作负载的经济高效的推理模型, images, and videos to generate text. 诺瓦 2 Lite demonstrates standout capabilities in processing...

The largest model in the Ministral 3 家庭, 部长级的 3 14B 提供可与更大的 Mistral Small 相媲美的前沿功能和性能 3.2 24B对应方. A powerful and efficient language...

A balanced model in the Ministral 3 家庭, 部长级的 3 8B 是一个强大的, 具有视觉功能的高效微型语言模型.

A balanced model in the Ministral 3 家庭, 部长级的 3 8B 是一个强大的, 具有视觉功能的高效微型语言模型.

The smallest model in the Ministral 3 家庭, 部长级的 3 3B 是一个强大的, 具有视觉功能的高效微型语言模型.

米斯特拉尔大号 3 2512 is Mistral’s most capable model to date, 采用具有 41B 个活动参数的稀疏专家混合架构 (675B总计), and released under the Apache 2.0 执照.

米斯特拉尔大号 3 2512 is Mistral’s most capable model to date, 采用具有 41B 个活动参数的稀疏专家混合架构 (675B总计), and released under the Apache 2.0 执照.

DeepSeek-V3.2 是一种大型语言模型，旨在协调高计算效率与强大的推理和代理工具使用性能. 它引入了 DeepSeek 稀疏注意力 (DSA), a fine-grained sparse attention mechanism...

Claude Opus 4.5 Anthropic 的前沿推理模型针对复杂的软件工程进行了优化, agentic workflows, 和长期使用电脑. 它提供强大的多式联运能力, competitive performance across real-world coding and...

Claude Opus 4.5 Anthropic 的前沿推理模型针对复杂的软件工程进行了优化, agentic workflows, 和长期使用电脑. 它提供强大的多式联运能力, competitive performance across real-world coding and...

Nano Banana Pro is Google’s most advanced image-generation and editing model, built on Gemini 3 Pro. It extends the original Nano Banana with significantly improved multimodal reasoning, real-world grounding, and...

GPT-5.1 is the latest frontier-grade model in the GPT-5 series, offering stronger general-purpose reasoning, improved instruction adherence, and a more natural conversational style compared to GPT-5. It uses adaptive reasoning...

GPT-5.1 is the latest frontier-grade model in the GPT-5 series, offering stronger general-purpose reasoning, improved instruction adherence, and a more natural conversational style compared to GPT-5. It uses adaptive reasoning...

GPT-5.1-Codex is a specialized version of GPT-5.1 optimized for software engineering and coding workflows. 它专为交互式开发会话和长时间的开发而设计。, independent execution of complex engineering tasks....

GPT-5.1-Codex-Mini is a smaller and faster version of GPT-5.1-Codex

Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning. Built on the trillion-parameter Mixture-of-Experts (MoE) architecture introduced in...

Amazon Nova Premier is the most capable of Amazon’s multimodal models for complex reasoning tasks and for use as the best teacher for distilling custom models.

Exclusively available on the OpenRouter API, Sonar Pro's new Pro Search mode is Perplexity's most advanced agentic search system. It is designed for deeper reasoning and analysis. Pricing is based...

Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance. It excels at speech transcription, translation and audio understanding. Input audio...

gpt-oss-safeguard-20b is a safety reasoning model from OpenAI built upon gpt-oss-20b. This open-weight, 21B 参数专家混合 (MoE) model offers lower latency for safety tasks like content classification, LLM filtering, and trust...

MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows. With 10 billion activated parameters (230 billion total), it delivers near-frontier intelligence across general reasoning,...

Qwen3-VL-32B-Instruct is a large-scale multimodal vision-language model designed for high-precision understanding and reasoning across text, images, and video. With 32 billion parameters, it combines deep visual perception with advanced text...

Granite-4.0-H-Micro is a 3B parameter from the Granite 4 family of models. These models are the latest in a series of models released by IBM. They are fine-tuned for long...