GPT-4.1 mini vs GPT-5 nano

GPT-5 nano is cheaper for every workload modelled here.

List prices checked

Specifications and rates

PropertyGPT-4.1 miniGPT-5 nano
Input per 1M$0.400$0.050
Output per 1M$1.60$0.400
Cached input per 1M$0.100$0.0050
Blended (3:1)$0.700$0.138
Context window1M400K
Max output33K128K
Inputs acceptedtext, imagetext, image

Source: OpenAI pricing (May 2026) · OpenAI pricing (May 2026). Published list prices only — committed-use discounts, regional variation, and per-request charges are not included. Not yet independently re-checked against the vendor’s page.

Monthly cost by workload

Identical token volumes through both rate cards. Output-heavy profiles are where the ranking flips.

WorkloadGPT-4.1 miniGPT-5 nanoCheaper
In-app chat assistantA support or help assistant handling a few thousand conversations a month, with short prompts and short answers.$16$3.00GPT-5 nanosaves $13/mo (81%)
RAG over documentsRetrieval-augmented answering where each request stuffs several retrieved passages into the prompt, so input dominates.$144$21GPT-5 nanosaves $123/mo (85%)
Coding agentAn agentic loop that reads files and writes patches, generating heavy output and re-sending large context on every turn.$320$60GPT-5 nanosaves $260/mo (81%)

What each one is for

GPT-4.1 mini

by OpenAI

The million-token context window at a fifth of the price, which is an unusual combination and the main reason this model is still in production stacks.

GPT-5 nano

by OpenAI

The cheapest model OpenAI publishes, intended for classification, routing, and cleanup passes where the task is simple and the volume is enormous.

Common questions

Is GPT-4.1 mini or GPT-5 nano cheaper in 2026?
GPT-5 nano, for every workload modelled here. On a blended 3:1 basis it costs $0.138 per million tokens against GPT-4.1 mini's $0.700.
Which has the larger context window?
GPT-4.1 mini, at 1M tokens against 400K. A larger window only helps if you can afford to fill it — the cost of one full-context request is in the table above.
Which is cheaper for output-heavy work like agents?
GPT-5 nano, at $0.400 per million output tokens against $1.60. Reasoning tokens bill as output, so for agentic loops this axis usually decides the invoice regardless of what the input rates look like.
What is not included in these figures?
Batch-processing discounts, fine-tuning, long-context surcharges above a provider's standard tier, and any enterprise agreement. These are published pay-as-you-go list prices, checked against each provider's own pricing page on the date shown.

Other model comparisons