AiCostCompare

GPT-5 Mini

by OpenAI · openai/gpt-5-mini · #147 cheapest of 325 paid models

Prices updated Jul 22, 2026, 1:41 AM UTC · refreshed hourly

GPT-5 Mini is a compact version of GPT-5, designed to handle lighter-weight reasoning tasks. It provides the same instruction-following and safety-tuning benefits as GPT-5, but with reduced latency and cost....

Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5Gemini 3.6 Flash

Pricing

Input

$0.25

per 1M tokens

Output

$2.00

per 1M tokens

Blended (3:1)

$0.688

3 input : 1 output

Cache read

$0.025

per 1M cached tokens

Cache write

per 1M tokens

Cache & batch economics

Effective prices for GPT-5 Mini with prompt caching.

Cache hit rateEffective blended $/1MExample request*vs no cache
0%$0.688$0.0275
50%$0.603$0.0163−41%
90%$0.536$0.00725−74%

*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates for this provider’s batch API. Confirm on the vendor’s pricing page.

What a request costs

WorkloadInput tokensOutput tokensCost / requestCost / 1K requests
Short chat message500300$0.000725$0.725
Document summary8,0001,000$0.004$4.00
Codebase question (RAG)30,0002,000$0.0115$11.50
Long-context analysis150,0005,000$0.0475$47.50

Performance

Intelligence Index

25.3

AA composite quality

Coding Index

15.6

Math Index

90.7

Agentic Index

19.4

Output speed

111 tok/s

median

Time to first token

87.43s

median

Time to first answer

87.43s

after reasoning tokens

Artificial Analysis benchmarks

BenchmarkScore
MMLU-Pro83.7%
GPQA Diamond82.8%
Humanity's Last Exam19.7%
LiveCodeBench83.8%
SciCode39.2%
AIME 202590.7%
IFBench75.4%
AA-LCR (Long Context Reasoning)68.0%
Terminal-Bench Hard33.3%
Terminal-Bench 2.13.7%
τ²-Bench (Telecom)68.4%
τ³-Bench Banking14.6%

Design Arena Elo

CategoryElo
Website1,152
UI components1,147
Data viz1,164
SVG1,138
3D1,096
Game dev1,187
Code categories1,148
ASCII art1,163

Specs

Context window

400K

tokens

Max output

128K

tokens

Input modalities

text, image, file

Tokenizer

GPT