OpenAI
GPT-5 pricing
OpenAI's flagship general-purpose model. 400K context window, 8x cheaper input than the o3 reasoning tier.
Input
$1.25/ 1M tok
Output
$10.00/ 1M tok
- Context window
- 400K
- Max output
- 128K
- Cached input
- $0.125 / 1M
- Verified
- 2026-04-06
GPT-5 sits at the centre of OpenAI's line-up: cheap enough to ship in production at scale, capable enough to handle most reasoning, code, and long-document tasks without falling back to the more expensive reasoning models. At $1.25 per 1M input and $10 per 1M output, the 8:1 output-to-input ratio is typical of frontier models - most of your bill will come from the response, not the prompt.
The 400K context window covers everything short of book-length documents, and the 128K max output is enough for any realistic single-call workflow. Use GPT-5 mini ($0.25 / $2.00) when you can - it's 5x cheaper for the same context window and handles most chat, summary, and routing tasks indistinguishably. Reach for the reasoning tier (o3 / o4-mini) only when you actually need chain-of-thought.
For pre-flight cost forecasts on your specific prompt, paste it into the estimator below - Calcis counts your input tokens with the same encoder OpenAI's billing uses (o200k_base) and predicts the response length from the prompt itself.
Estimate your cost on GPT-5
Paste your prompt into the estimator, pick GPT-5, and see the exact dollar cost - input tokens counted with the provider's own tokenizer, output tokens predicted by our regression model.
Frequently asked
- How much does GPT-5 cost per request?
- It depends on your prompt and the response length. A 1,000-token prompt with a typical 500-token reply costs about $0.0064 ($0.00125 input + $0.005 output). Calcis estimates your specific cost without any guesswork.
- Is GPT-5 cheaper than Claude Sonnet 4.6?
- On input, yes - $1.25 vs $3.00 per 1M tokens. On output, GPT-5 is also cheaper ($10 vs $15 per 1M). Sonnet 4.6 has a 1M-token context window vs GPT-5's 400K. Pick on capability and context size, not just price.
- Does GPT-5 have a long-context surcharge?
- No. Unlike Gemini 2.5 Pro, OpenAI doesn't double-bill above a threshold - the same per-token rates apply across the full 400K context window.
- What's the cached input price for GPT-5?
- $0.125 per 1M tokens - a 90% discount on the standard input rate. OpenAI applies prompt caching automatically when prefixes repeat across calls within a 5-minute window.
Pricing verified 2026-04-06 from the provider's rate card.