OpenAI API pricing
31 models tracked · headquartered in US · official pricing
Free tier: No free tier What OpenAI gives you free, in its own words
| Model | Input | Output | Context | |
|---|---|---|---|---|
| GPT-5 Nano | $0.05 | $0.4 | 272K | reasoning vision |
| GPT-6 Luna | $0.1 | $0.5 | 922K | reasoning vision |
| GPT-4o Mini legacy | $0.15 | $0.6 | 128K | vision |
| GPT-5.6 Luna | $0.2 | $1.2 | 922K | reasoning vision |
| GPT-5.4 Nano | $0.2 | $1.25 | 272K | reasoning vision |
| GPT-5 Mini | $0.25 | $2 | 272K | reasoning vision |
| GPT-4.1 Mini legacy | $0.4 | $1.6 | 1.05M | vision |
| GPT-5.4 Mini | $0.75 | $4.5 | 272K | reasoning vision |
| o3-mini retired | $1.1 | $4.4 | 200K | reasoning vision |
| o4-mini retired | $1.1 | $4.4 | 200K | reasoning vision |
| GPT-6.1 Terra | $1 | $5 | 922K | reasoning vision |
| GPT-5 | $1.25 | $10 | 272K | reasoning vision |
| GPT-5.1 | $1.25 | $10 | 272K | reasoning vision |
| GPT-4.1 legacy | $2 | $8 | 1.05M | vision |
| o3 | $2 | $8 | 200K | reasoning vision |
| GPT-6 Sol | $2 | $10 | 922K | reasoning vision |
| GPT-6.1 Sol | $2 | $10 | 922K | reasoning vision |
| GPT-4o legacy | $2.5 | $10 | 128K | vision |
| GPT-5.6 Terra | $2 | $12 | 922K | reasoning vision |
| GPT-5.2 | $1.75 | $14 | 272K | reasoning vision |
| GPT-5.4 | $2.5 | $15 | 1.05M | reasoning vision |
| GPT-5.5 | $5 | $30 | 1.05M | reasoning vision |
| GPT-5.6 Sol | $4 (Promotional rate until 2026-11-21 — standard rate $5/M) | $20 (Promotional rate until 2026-11-21 — standard rate $30/M) | 922K | reasoning vision |
| GPT-6 Astra | $10 | $50 | 922K | reasoning vision |
| o1 retired | $15 | $60 | 200K | reasoning vision |
| o3-pro legacy | $20 | $80 | 200K | reasoning vision |
| GPT-5 Pro legacy | $15 | $120 | 400K | reasoning vision |
| GPT-5.2 Pro | $21 | $168 | 272K | reasoning vision |
| GPT-5.4 Pro | $30 | $180 | 1.05M | reasoning vision |
| GPT-5.5 Pro | $30 | $180 | 1.05M | reasoning vision |
| o1-pro retired | $150 | $600 | 200K | reasoning vision |
Compare OpenAI’s models on your own workload
How OpenAI prices its models
Output prices across OpenAI's 21 current models run from $0.4 per 1M tokens (GPT-5 Nano) to $180 (GPT-5.5 Pro), a 450x spread. Picking the smallest model that does the job is usually the biggest saving available.
Discounts: caching and batch
18 of 21 current models list a cached-input rate, from 90% off the fresh input rate on GPT-5 Nano. 15 list a batch discount for work that can wait.
Recorded price changes
No vendor price change to a OpenAI input or output rate is on record since tracking began. Corrections we made to our own readings are listed on the changes page.
Where these prices come from
Every rate above is read from OpenAI's own pricing page (linked here) and re-checked every morning. A disputed figure is marked rather than hidden; see how PromptSpend checks prices.
Common questions
How much do OpenAI API models cost?
PromptSpend tracks 31 OpenAI models. The cheapest on a blended rate is GPT-5 Nano, at $0.05 per 1M input tokens and $0.4 per 1M output tokens; the most expensive is o1-pro, at $150 in and $600 out.
Does OpenAI have a free API tier?
PromptSpend's verdict is "No free tier". Its free-tier page gives OpenAI's own wording for each claim, with a link to the page it came from and the date it was read.
Where do the OpenAI prices come from?
From OpenAI's own pricing page, https://developers.openai.com/api/docs/pricing, with each model's check date shown on its page. Prices change without notice, and regional premiums, priority tiers and negotiated discounts are not included.