AI Model Lineup

Offering diverse AI model options to meet different scenarios and performance needs

Standard Series

GPT-5.4-mini

Lightweight and fast version of GPT-5.4, delivering the core capabilities of the GPT-5 series at a highly competitive price. Faster responses and lower cost make it ideal for latency-sensitive or high-frequency use cases like everyday coding assistance and conversational tasks.

Supported Depths:LowMediumHigh
Input$0.75/M·Cached Input$0.075/M·Output$4.50/M

GPT-5.4

The flagship new version of the GPT-5 series, achieving comprehensive breakthroughs in multimodal understanding, long-form reasoning, and complex task planning. Supports four reasoning depths — Low, Medium, High, and xhigh — covering all scenarios from everyday conversation to advanced research.

Supported Depths:LowMediumHighxhigh
≤272KInput$2.50/M·Cached Input$0.25/M·Output$15.00/M
>272KInput$5.00/M·Cached Input$0.50/M·Output$22.50/M

GPT-5.2

Latest version of the GPT-5 series with knowledge cutoff updated to August 31, 2025. Features a 400k token context window and 128k maximum output tokens, reaching new heights in reasoning capability and performance.

Supported Depths:LowMediumHighxhigh
Input$1.75/M·Cached Input$0.18/M·Output$14.00/M

GPT-5.1

Enhanced version of GPT-5 with significantly improved performance and response speed. Optimized context understanding and multi-turn conversation capabilities provide users with a smoother interactive experience.

Supported Depths:LowMediumHigh
Input$1.25/M·Cached Input$0.125/M·Output$10.00/M

GPT-5

OpenAI's fifth-generation large language model, achieving major breakthroughs in comprehension, reasoning, and creativity. Ideal for complex task processing and advanced conversational scenarios.

Supported Depths:LowMediumHigh
Input$1.25/M·Cached Input$0.125/M·Output$10.00/M

Codex Code-Specialized Series

GPT-5.3-Codex

The latest upgrade in the Codex series, built upon GPT-5.2-Codex with further optimizations. Delivers significant improvements in code comprehension depth, multi-file collaborative editing, long-context code reasoning, and automated debugging, providing more powerful intelligent programming support for large-scale software engineering projects.

Supported Depths:LowMediumHighxhigh
Input$1.75/M·Cached Input$0.18/M·Output$14.00/M

GPT-5.2-Codex

Code-specialized model built for demanding software engineering and defensive cybersecurity tasks, released on December 18, 2025. Compared to GPT-5.1-Codex, it offers significant improvements in context compaction, massive code refactors and migrations, Windows environment compatibility, and cybersecurity defense capabilities, setting a new industry benchmark for agentic coding.

Supported Depths:LowMediumHighxhigh
Input$1.75/M·Cached Input$0.18/M·Output$14.00/M

GPT-5.1-Codex

Upgraded code-specialized model with further improvements in code understanding and generation quality. Supports more programming languages and frameworks, with deeper understanding of complex project structures and architectural patterns.

Supported Depths:LowMediumHigh
Input$1.25/M·Cached Input$0.125/M·Output$10.00/M

GPT-5-Codex

GPT-5 version optimized specifically for code generation and programming tasks. Excels in code completion, debugging, refactoring, and technical documentation writing, serving as a powerful assistant for developers.

Supported Depths:LowMediumHigh
Input$1.25/M·Cached Input$0.125/M·Output$10.00/M

Codex Mini Lightweight Series

GPT-5.1-Codex-Mini

Optimized lightweight code model combining GPT-5.1's performance improvements with Mini series efficiency. Achieves ideal balance between speed and quality, perfect for daily development work.

Supported Depths:LowMediumHigh
Input$1.25/M·Cached Input$0.125/M·Output$10.00/M

GPT-5-Codex-Mini

Lightweight code model that maintains high-quality code generation while significantly improving response speed and performance efficiency. Ideal for quick code completion and real-time programming assistance scenarios.

Supported Depths:LowMediumHigh
Input$1.25/M·Cached Input$0.125/M·Output$10.00/M

Codex Max Flagship Series

GPT-5.1-Codex-Max

Flagship code model providing the most powerful code understanding and generation capabilities. Supports extra-high depth reasoning and can handle the most complex software engineering tasks, including large-scale system architecture design and performance optimization.

Supported Depths:LowMediumHighxhigh
Input$1.25/M·Cached Input$0.125/M·Output$10.00/M

Claude Code Series

Use these models directly inside Claude Code via protocol format conversion at lower cost. See thesetup guide

cc-gpt-5.4

Claude Code-compatible mapping of GPT-5.4. Via Codex → Claude Code protocol conversion, you can call it directly inside the Claude Code client and enjoy the full capabilities of GPT-5.4 at the same pricing.

Supported Depths:LowMediumHighxhigh
≤272KInput$2.50/M·Cached Input$0.25/M·Output$15.00/M
>272KInput$5.00/M·Cached Input$0.50/M·Output$22.50/M
支持的模型 - CodexZH | GPT-5, GPT-5.1, GPT-5.2, Codex 全系列模型