Loading page
LightweightOpenAI
GPT-5 Nano
GPT-5 Nano is OpenAI's lightweight model in the GPT-5 family, designed for high-volume applications with a 128K token context window.
Context 128K
Tier Lightweight
Knowledge Dec 2024
Tools Supported
Input from
$0.025 / 1M tokens
across 3 providers
API Pricing
Cheapest on OpenAI — 38% below avg| Provider | Input / 1M | Output / 1M | Cached / 1M |
|---|---|---|---|
| $0.025 | $0.200 | $0.0025 | |
| $0.025 | $0.200 | $0.0025 | |
| $0.050 | $0.400 | $0.0050 | |
| $0.050 | $0.400 | $0.0050 | |
| $0.050 | $0.400 | $0.0050 |
Prices updated daily. Last check: Sep 6, 2026
GPT-5 Nano pricing by provider
Input, output, and batch rates, plus alternatives, for one provider at a time.
Performance & Benchmarks
Source: Artificial Analysis →Intelligence
13.7 / 100
Math
83.7 / 100
Reasoning & Knowledge
- MMLU-Pro78.0%
- GPQA Diamond67.6%
- Humanity's Last Exam9.5%
Coding
- LiveCodeBench78.9%
Math
- AIME 202583.7%
Agentic & Tool Use
- Terminal-Bench Hard12.1%
- τ²-bench36.5%
Instruction & Long Context
- IFBench67.6%
- Long-Context Reasoning45.0%
Benchmarks measured Sep 2026. Scores are independent evaluations, not vendor-reported.
Model Details
General
- Creator
- OpenAI
- Family
- GPT
- Tier
- Lightweight
- Context Window
- 128K
- Knowledge Cutoff
- Dec 2024
- Modalities
- Text
Capabilities
- Tool Calling
- Yes
- Open Source
- No
- Subtypes
- Chat Completion
Strengths & Limitations
Strengths
- 128K token context window supports processing of large documents
- Tool calling functionality enables API integrations and structured outputs
- December 2024 knowledge cutoff provides relatively current information
- Lightweight tier optimized for high-throughput applications
- Part of GPT-5 generation with underlying architectural improvements
- Chat completion interface supports conversational workflows
- Maintained by OpenAI with enterprise-grade API infrastructure
Limitations
- Text-only modality lacks image, audio, or video input support
- Proprietary model with no access to weights or local deployment
- Lightweight tier may have reduced reasoning capabilities compared to GPT-5 Turbo or GPT-5
- No multimodal capabilities limit use in vision or audio applications
- Smaller context window than some competing models in 2026
Key Features
•128K token context window
•Text-based chat completion
•Tool calling with external API integration
•Streaming response support
•JSON mode for structured outputs
•December 2024 knowledge cutoff
•Batch processing capabilities
•Function calling with parameters
About GPT-5 Nano
GPT-5 Nano is OpenAI's lightweight tier model within the GPT-5 family, positioned as a cost-effective option for applications requiring high throughput and efficiency. As part of OpenAI's latest generation, it sits below the more capable GPT-5 and GPT-5 Turbo variants in terms of reasoning complexity but maintains strong performance for straightforward language tasks.
The model supports text-only interactions through chat completion with a 128K token context window, allowing it to process substantial documents and maintain extended conversations. GPT-5 Nano includes tool calling capabilities, enabling it to interact with external APIs and structured data sources. With a knowledge cutoff of December 2024, it has access to relatively recent information compared to earlier model generations.
GPT-5 Nano targets use cases where speed and cost efficiency take priority over maximum reasoning capability. It competes with other lightweight models in scenarios requiring reliable text generation, basic analysis, and automated responses at scale, while still benefiting from the architectural improvements present in the GPT-5 generation.
Common Use Cases
GPT-5 Nano is well-suited for high-volume applications requiring reliable text processing without the computational overhead of flagship models. Its lightweight design makes it effective for customer service automation, content moderation, basic document analysis, and simple reasoning tasks. The 128K context window supports applications like document summarization, FAQ generation, and extended conversation handling. With tool calling capabilities, it can power chatbots that need to query databases, call APIs, or generate structured responses. Organizations benefit from using GPT-5 Nano for workflows requiring consistent quality at scale, such as email drafting, content classification, basic coding assistance, and routine analytical tasks where maximum reasoning depth is not critical.
Frequently Asked Questions
How much does GPT-5 Nano cost per million tokens?
GPT-5 Nano pricing varies by provider and pricing type (standard vs batch). Check the pricing table above for current rates across all providers offering this model.
What is GPT-5 Nano best used for?
GPT-5 Nano excels at high-volume text processing tasks like customer service automation, document summarization, content moderation, and basic reasoning workflows. Its lightweight design and 128K context window make it cost-effective for applications requiring reliable performance without the computational requirements of larger models.
How does GPT-5 Nano compare to other models in the GPT-5 family?
GPT-5 Nano is the lightweight tier option, optimized for speed and cost efficiency rather than maximum reasoning capability. While it shares the GPT-5 generation's architectural improvements and maintains tool calling functionality, it trades some reasoning depth for better throughput compared to GPT-5 Turbo and the flagship GPT-5 model.