Reasoning · 1 checks · max 800 tokens · benched 2026-09-16
OpenAI models side by side on "Stacked discounts and tax": GPT-5.4 Nano scores 10/10; GPT-5.4 Nano is the cheapest answer scoring 8+ at $0.08 per 1,000 runs. Outputs, checks, judge reasons, latency and cost.
System
Think carefully, then give only the final answer in the requested format.
User
An item costs $250.00. A 20% discount is applied, then a $15.00 coupon is subtracted, and finally 8% sales tax is added to the resulting amount. What is the final price? Reply with only the amount in dollars to two decimal places, for example 12.34, with no currency symbol.
Rubric for the judge: The single correct answer 199.80 (250 x 0.8 = 200; 200 - 15 = 185; 185 x 1.08).
Every cell is one OpenRouter call at temperature 0 with the prompt's token cap and reasoning effort "low" where the model supports it. Cost is usage × the catalogue rate in models.json. Quality is one judge call to anthropic/claude-haiku-4.5 against the prompt's rubric, cached per prompt version.
openai/gpt-5.5
95 in · 46 out (37 reasoning) · 1 words · checks 1/1
199.80
Judge: The output matches the correct answer of 199.80 in the exact format requested (two decimal places, no currency symbol).
openai/gpt-5.4
95 in · 44 out (35 reasoning) · 1 words · checks 1/1
199.80
Judge: The output matches the correct answer of 199.80 in the exact format requested (two decimal places, no currency symbol).
openai/gpt-5.4-mini
95 in · 44 out (35 reasoning) · 1 words · checks 1/1
199.80
Judge: The output matches the correct answer of 199.80 in the exact format requested (two decimal places, no currency symbol).
openai/gpt-5.4-nano
95 in · 51 out (42 reasoning) · 1 words · checks 1/1
199.80
Judge: The output matches the correct answer of 199.80 in the exact format requested (two decimal places, no currency symbol).
openai/gpt-5.3-codex
95 in · 46 out (37 reasoning) · 1 words · checks 1/1
199.80
Judge: The output matches the correct answer of 199.80 in the exact format requested (two decimal places, no currency symbol).
Reasoning: The single correct answer 199.80 (250 x 0.8 = 200; 200 - 15 = 185; 185 x 1.08). The deterministic checks are exact.
If the judge's bar of 8/10 is good enough for you, GPT-5.4 Nano at $0.08 per 1,000 runs. If you need the top score, GPT-5.4 Nano at $0.08 per 1,000 runs.
Run it yourself
Re-run this prompt on OpenAI with your own OpenRouter key, or tweak the wording and see what changes.
More OpenAI prompts
If our calculators helped you cut down on hidden AI wallet leaks, thanks for using them. A tiny fraction of your savings is what keeps our pricing indexes updated daily.
Not sure which AI is cheapest for your use case? Find out in 30 seconds — no signup required.
AI cost intelligence
Join the OverpayingForAI list for pricing updates, cheaper alternatives, and practical buying guidance.
Now tracking 50+ AI tools, models, platforms, subscriptions, coding tools, and automation products.