Loading page
FlagshipOpenAI
GPT-5
GPT-5 is OpenAI's flagship multimodal model offering advanced reasoning, coding, and vision capabilities with a 200K token context window.
Context 200K
Tier Flagship
Knowledge Dec 2024
Tools Supported
Modalities text, image
Input from
$0.625 / 1M tokens
across 3 providers
API Pricing
Cheapest on OpenAI — 38% below avg| Provider | Input / 1M | Output / 1M | Cached / 1M |
|---|---|---|---|
| $0.625 | $5.00 | $0.063 | |
| $0.625 | $5.00 | $0.063 | |
| $1.25 | $10.00 | $0.125 | |
| $1.25 | $10.00 | $0.125 | |
| $1.25 | $10.00 | $0.125 |
Prices updated daily. Last check: Sep 8, 2026
GPT-5 pricing by provider
Input, output, and batch rates, plus alternatives, for one provider at a time.
Performance & Benchmarks
Source: Artificial Analysis →Intelligence
27.1 / 100
Coding
37.8 / 100
Math
94.3 / 100
Reasoning & Knowledge
- MMLU-Pro87.1%
- GPQA Diamond85.4%
- Humanity's Last Exam28.5%
Coding
- LiveCodeBench84.6%
Math
- AIME 202594.3%
- AIME95.7%
- MATH-50099.4%
Agentic & Tool Use
- Terminal-Bench Hard32.6%
- Terminal-Bench v2.135.2%
- τ²-bench84.8%
- τ-bench Banking22.1%
Instruction & Long Context
- IFBench73.1%
- Long-Context Reasoning78.2%
Benchmarks measured Sep 2026. Scores are independent evaluations, not vendor-reported.
Model Details
General
- Creator
- OpenAI
- Family
- GPT
- Tier
- Flagship
- Context Window
- 200K
- Knowledge Cutoff
- Dec 2024
- Modalities
- Text, Image
Capabilities
- Tool Calling
- Yes
- Open Source
- No
- Subtypes
- Chat Completion, Code Generation
Strengths & Limitations
Strengths
- 200,000 token context window enables processing of lengthy documents and conversations
- Multimodal support with both text and image input capabilities
- Tool calling functionality for integration with external APIs and functions
- Strong code generation performance across multiple programming languages
- Recent knowledge cutoff of December 2024 provides up-to-date information
- Output speed of 68.28 tokens per second for responsive interactions
- Advanced reasoning capabilities for complex problem-solving tasks
Limitations
- Proprietary model with no open-source weights available
- Time to first token of 9.457 seconds may impact real-time applications
- No audio input or output modalities supported
- Higher computational requirements typical of flagship-tier models
- Limited to OpenAI's API ecosystem and pricing structure
Key Features
•200K token context window
•Text and image input processing
•Tool calling with function execution
•Chat completion API
•Code generation and analysis
•Streaming response support
•JSON mode for structured outputs
•Vision analysis and description
About GPT-5
GPT-5 is OpenAI's flagship large language model, representing the latest generation in the GPT family. As OpenAI's most capable model, it sits at the top of their model hierarchy, designed for complex reasoning, advanced coding tasks, and multimodal applications requiring sophisticated understanding.
The model features a 200,000 token context window and supports both text and image inputs, enabling comprehensive document analysis and visual reasoning tasks. GPT-5 includes tool calling capabilities and demonstrates strong performance in code generation and chat completion scenarios. With a knowledge cutoff of December 2024, it incorporates relatively recent training data compared to earlier model generations.
GPT-5 competes directly with other flagship models like Claude Opus 4.6 and Gemini 3.1 Pro in the premium tier market. Its combination of multimodal capabilities, large context window, and advanced reasoning makes it suitable for enterprise applications, complex research tasks, and sophisticated AI agent implementations where maximum capability is prioritized over cost efficiency.
Common Use Cases
GPT-5 is designed for demanding applications that require OpenAI's most advanced capabilities. Its large context window and multimodal support make it well-suited for complex document analysis, research assistance, and comprehensive content creation tasks. The model excels in sophisticated coding projects, technical writing, and multi-step reasoning problems. Organizations use GPT-5 for AI agent implementations, advanced customer support systems, and research applications where the combination of text and vision understanding is crucial. Its tool calling capabilities enable integration into complex workflows and business process automation where maximum model capability justifies the premium positioning.
Frequently Asked Questions
How much does GPT-5 cost per million tokens?
GPT-5 pricing varies by provider and usage type (standard vs batch processing). Check the pricing table above for current rates across all available providers offering GPT-5 access.
What is GPT-5 best used for?
GPT-5 excels in complex reasoning tasks, advanced code generation, multimodal analysis combining text and images, and applications requiring the full 200K token context window. It's optimal for AI agents, research assistance, comprehensive document analysis, and enterprise applications where maximum capability is prioritized.
How does GPT-5 compare to GPT-4 models?
GPT-5 represents OpenAI's latest flagship model with enhanced reasoning capabilities, more recent training data (December 2024 cutoff), and improved performance across coding and multimodal tasks. It maintains the same 200K context window as GPT-4 Turbo while offering faster output generation at 68.28 tokens per second.