Anthropic's streamlined hybrid model, autonomously navigates codebases, balancing performance and efficiency. Suitable for various development scenarios.
NEW Claude Opus 4
Anthropic's latest hybrid model, supports long-running agent workflows and persistent execution of complex tasks.
Gemini 2.5 Pro Preview
Preview of Google's latest reasoning model, excels at programming and science.
HOT Claude 3.7 Sonnet
Anthropic's hybrid model that autonomously decides when reasoning is needed, with outstanding coding abilities.
GPT-4o
A well-rounded and classic OpenAI model with strong multimodal understanding.
GPT‑4o Image
OpenAI’s advanced image generator, allowing continuous modifications and detail adjustments.
HOT Midjourney
The most popular and classic image model, supporting multiple parameter adjustments.
Runway
A video generation model known for excellent dynamic control and animation.
AI Detector
An AI content detection tool that accurately identifies AI-generated content in text, images, audio, and video.
OpenAI's latest lightweight model, improved coding abilities, and fast response speed.
Grok 3 Beta
Beta version of xAI's reasoning model by Elon Musk, with enhanced coding and financial quantification capabilities.
Llama 4 Maverick
Meta's latest open-source model, enhanced with improved multimodal recognition.
Llama 4 Scout
Meta's latest lightweight open-source model, focused on data integration and fact-checking.
Open Manus
A universal AI Agent that connects thought and action.
DeepSeek V3 0324
A highly cost-effective, fast and affordable conversational model.
Deep Research
An advanced self-developed Agent for searching online information, reasoning, verification, and organizing multi-step research explorations.
GPT-4o mini
OpenAI's most cost-effective and fast-responding model.
GPT-4.5 Preview
Open AI's experimental research model with an expanded knowledge base and verbose output.
DeepSeek R1
DeepSeek's classic reasoning model, excels at solving math problems and programming analysis.
Ideogram
A text-to-image model with strong instruction comprehension, producing accurate and efficient results.
Luma
A text-to-video model with precise text recognition and smooth camera movement.
FLUX
The open-source version of Runway's creative text-to-image model, offering creative freedom.
Chat
NEW Claude Sonnet 4
Anthropic's streamlined hybrid model, autonomously navigates codebases, balancing performance and efficiency. Suitable for various development scenarios.
NEW Claude Opus 4
Anthropic's latest hybrid model, supports long-running agent workflows and persistent execution of complex tasks.
NEW Saba
A Mistral model designed for the Middle East and Southeast Asia, supporting multiple regional languages and providing culturally relevant and accurate responses.
NEW Codestral 2501
Mistral's professional programming model with low latency and high efficiency, excelling at code completion, correction, and test generation.
NEW Qwen2.5 VL 32B Instruct
Alibaba's enhanced multimodal model, excels at visual analysis and mathematical reasoning.
NEW Qwen VL Plus
Alibaba's enhanced multimodal model, supports ultra-high resolution and excels at detail and text recognition.
NEW Qwen VL Max
An enhanced multimodal model by Alibaba, supporting complex visual tasks.
Gemini 2.5 Pro Preview
Preview of Google's latest reasoning model, excels at programming and science.
Phi 4
A highly efficient reasoning model by Microsoft, trained on high-quality datasets.
WizardLM-2 8x22B
An open-source model from Microsoft with strong comprehension of complex instructions.
GPT-4.1 Nano
OpenAI's lightest intelligent model, low-latency and fast response, ideal for classification and autocomplete tasks.
o3 Mini High
Based on o3 mini, now with added multimodal and online search capabilities.
o4 Mini High
Based on o4 mini, with optimized reasoning speed.
o4 Mini
A streamlined reasoning model with more multimodal features, built in a more compact way than o3 mini.
Grok 3 Mini Beta
Elon Musk's xAI lightweight reasoning model, with a reduced knowledge base and more precise expression.
Qwen3 30B A3B
Alibaba's open-source inference model with efficient activated parameters and outstanding inference capability.
Qwen3 14B
Alibaba's open-source inference model, excels in reasoning, programming, and multilingual processing.
Qwen3 32B
Alibaba's open-source inference model, excels at complex reasoning.
Qwen3 235B A22B
Alibaba's open-source inference model, a large-scale Mixture of Experts model.
Gemini 2.5 Flash Preview
A fast version of Google's latest reasoning model, excels at mathematics and coding with quick responses.