← Back to Calculator

GPT-5.6 Luna vs GPT-5.4 nano

Detailed comparison of API pricing, technical specifications, and benchmark scores. Updated 2026-06-29.

Cost at a Glance

GPT-5.6 Luna is 4.9× more expensive than GPT-5.4 nano — switching could save 79.4% on your monthly API bill.

At 5M input + 2M output tokens per month:

GPT-5.6 Luna$17.00/movsGPT-5.4 nano$3.50/mo

Choosing GPT-5.4 nano could save you $13.50/mo

Price Comparison (per 1M tokens)

Input (per 1M tokens)
$1.00vs$0.20
Output (per 1M tokens)
$6.00vs$1.25
Cache Write (per 1M tokens)
$1.25vs$0.25
Cache Read (per 1M tokens)
$0.10vs$0.02
Batch Input (per 1M tokens)
$0.50vs$0.10
Batch Output (per 1M tokens)
$3.00vs$0.63

Prices are per 1 million tokens. Actual costs depend on your monthly token volume. Use the calculator →

Specifications

Context Window
1Mvs400K
Max Output Tokens
64Kvs128K
Knowledge Cutoff
2026-04vs2025-08
Modalities
TextvsText, Image
Prompt Caching
YesvsYes
Batch API
YesvsYes
Thinking / Reasoning
NovsYes

Benchmark Scores

SWE-bench Verified
62.5vs65.0
MMLU-Pro
79.0vs74.0
HumanEval
89.5vs85.0

Benchmark scores from official provider reports. Higher is better. Missing scores indicate the provider has not published that benchmark yet.

Use Cases

GPT-5.6 Luna

fastcheapgeneral

GPT-5.4 nano

cheapestfastsimple

About the Providers

GPT-5.6 Luna is developed by OpenAI. GPT-5.4 nano is developed by OpenAI. Both models are from the same provider, representing different tiers in the OpenAI model lineup. For the most up-to-date pricing, always check the official API documentation from each provider.

Compare More AI Models