models/openai/gpt-5-nano
O
OpenAI·active

GPT-5 Nano

OpenAI's efficiency model. Context window: 400K tokens.

Overall score
4.08
/5.00 · ranked #50
Input
$0.050
per 1M tokens
Output
$0.400
per 1M tokens
Context
400K
tokens
Blended
$0.313
3:1 out:in ratio

Price drops, new benchmarks, model updates. Stay current on GPT-5 Nano.

One email per change. Unsubscribe anytime.

modelpicker.aipowered by live benchmark data

Scores by test

Methodology →
Structured Output
5.0
Strategic Analysis
4.0
Constrained Rewriting
3.0
Creative Problem Solving
3.0
Tool Calling
4.0
Faithfulness
4.0
Classification
3.0
Long Context
5.0
Safety Calibration
4.0
Persona Consistency
4.0
Agentic Planning
4.0
Multilingual
5.0
Tabular Data
5.0
MATH Level 5
95.2
AIME 2025
81.1

What you need to know

GPT-5 Nano is a high-efficiency model optimized for structured data and massive context. Its primary technical advantage is the combination of a 400K context window and perfect scores in structured output, tabular data, and multilingual processing. This makes it highly effective for parsing large, diverse datasets into precise formats.

The model demonstrates exceptional quantitative reasoning, scoring 95.2% on MATH Level 5 and 81.1% on AIME 2025. These results are disproportionately high compared to its overall rank of 43rd among 71 models, indicating a specialized strength in mathematics that outweighs its general-purpose performance.

At a blended cost of $0.313/MTok, the model is priced for high-volume automation. While it excels at logic and structure, it struggles with nuanced linguistic tasks, specifically classification, constrained rewriting, and creative problem solving, where it only achieves a 3/5 internal rating.

Use this model if your workflow requires high-accuracy mathematical reasoning, large-scale document analysis, or strict adherence to structured output schemas. Skip this model if your application relies on creative synthesis or precise text classification.

Strengths — Top 3

Structured Output5.0/5.0
Long Context5.0/5.0
Multilingual5.0/5.0

Relative weaknesses — Bottom 3

Constrained Rewriting3.0/5.0
Creative Problem Solving3.0/5.0
Classification3.0/5.0

Similar models

QQwen: Qwen3.5-9B$0.1224.31NNVIDIA: Nemotron 3 Nano 30B A3B$0.1634.08GGemini 3.1 Flash Lite Preview$1.194.46GGemini 2.5 Flash$1.954.15