Is GPT-5 or GPT-5 Mini cheaper?

For a typical request (10,000 input + 2,000 output tokens), GPT-5 Mini is cheaper — about 80% less, or roughly $2600 saved per 100,000 requests. GPT-5 runs $1.25/$10 per 1M input/output tokens; GPT-5 Mini runs $0.25/$2.

Which has the larger context window?

Both support a 400,000-token context window.

How accurate are these token counts?

GPT-5: Exact tokenization via the canonical OpenAI vocab (o200k_base). GPT-5 Mini: Exact tokenization via the canonical OpenAI vocab (o200k_base). The dollar math itself is exact once the token count is known.

GPT-5 vs GPT-5 Mini: pricing & cost comparison

On input tokens, GPT-5 Mini is the cheaper of the two — 80% less per million ($1.25 vs $0.25). On output, GPT-5 Mini is 80% cheaper ($10 vs $2) — and since output is usually the dominant cost driver, that gap matters more than it looks.

Side by side

	GPT-5	GPT-5 Mini
Input / 1M tokens	$1.25	$0.25
Output / 1M tokens	$10	$2
Context window	400,000	400,000
Token-count accuracy	exact	exact
Cost — 10,000 input + 2,000 output tokens	$0.0325	$0.0065

What a real request costs

Take a representative turn — 10,000 input + 2,000 output tokens. GPT-5 comes to $0.0325, GPT-5 Mini to $0.0065. Across 100,000 requests that's a $2600 swing in favour of GPT-5 Mini. To run the numbers on your actual prompt, paste it into the calculator and toggle Compare across all models.

Same family — pick on price

Moving between GPT-5 and GPT-5 Mini is a one-line model-string change — same family, same tokenizer, nothing to re-encode. That makes this a straight cost/quality call: send the routine bulk of traffic to GPT-5 Mini ($0.25/$2 per 1M) and reserve GPT-5 for the requests that visibly need its extra headroom. On the worked example above that split is worth about 80% per request — small per call, real money once you multiply by volume.

See the full breakdown on the dedicated pages for GPT-5 and GPT-5 Mini.

FAQ

Is GPT-5 or GPT-5 Mini cheaper?: For a typical request (10,000 input + 2,000 output tokens), GPT-5 Mini is cheaper — about 80% less, or roughly $2600 saved per 100,000 requests. GPT-5 runs $1.25/$10 per 1M input/output tokens; GPT-5 Mini runs $0.25/$2.
Which has the larger context window?: Both support a 400,000-token context window.
How accurate are these token counts?: GPT-5: Exact tokenization via the canonical OpenAI vocab (o200k_base). GPT-5 Mini: Exact tokenization via the canonical OpenAI vocab (o200k_base). The dollar math itself is exact once the token count is known.

Both prices are computed from tokenmath's verified pricing table. Rates sourced from openai.com, verified 2026-05-09. Vendor pricing changes often — confirm before you commit.