API pricing · 32 models
OpenAI API pricing
What every OpenAI model costs per 1M tokens through the API, with the extra charges and discounts that change the bill.
Prices from OpenAI's pricing page as of
Prices per 1M tokens
US dollars, standard processing. The last column is what one request with 1,000 input and 500 output tokens costs; try your own volumes in the cost calculator.
| Model | Input | Cached input | Cache write | Output | Batch in | Batch out | 1K in + 500 out |
|---|---|---|---|---|---|---|---|
| GPT-6 | |||||||
| GPT-6 Astragpt-6-astra | $10.00 | $1.00 | $12.50 | $50.00 | $5.00 | $25.00 | $0.035 |
| GPT-6.1 Solgpt-6.1-sol | $2.00 | $0.10 | $2.50 | $10.00 | $1.00 | $5.00 | $0.007 |
| GPT-6 Solgpt-6-sol | $2.00 | $0.20 | $2.50 | $10.00 | $1.00 | $5.00 | $0.007 |
| GPT-6 Lunagpt-6-luna | $0.10 | $0.01 | $0.125 | $0.50 | $0.05 | $0.25 | $0.00035 |
| GPT-5.6 | |||||||
| GPT-5.6 Solpromo pricegpt-5.6-sol | $4.00 | $0.40 | $5.00 | $20.00 | — | — | $0.014 |
| GPT-5.6 Terragpt-5.6-terra | $2.00 | $0.20 | $2.50 | $12.00 | — | — | $0.008 |
| GPT-5.6 Lunagpt-5.6-luna | $0.20 | $0.02 | $0.25 | $1.20 | — | — | $0.0008 |
| GPT-5.5 | |||||||
| GPT-5.5gpt-5.5 | $5.00 | $0.50 | — | $30.00 | — | — | $0.02 |
| GPT-5.5 Progpt-5.5-pro | $30.00 | — | — | $180.00 | — | — | $0.12 |
| GPT-5.4 | |||||||
| GPT-5.4gpt-5.4 | $2.50 | $0.25 | — | $15.00 | — | — | $0.01 |
| GPT-5.4 Progpt-5.4-pro | $30.00 | — | — | $180.00 | — | — | $0.12 |
| GPT-5.4 minigpt-5.4-mini | $0.75 | $0.075 | — | $4.50 | — | — | $0.003 |
| GPT-5.4 nanogpt-5.4-nano | $0.20 | $0.02 | — | $1.25 | — | — | $0.00083 |
| GPT-5.3 and 5.2 | |||||||
| GPT-5.3 Codexgpt-5.3-codex | $1.75 | $0.175 | — | $14.00 | — | — | $0.0088 |
| GPT-5.2gpt-5.2 | $1.75 | $0.175 | — | $14.00 | — | — | $0.0088 |
| GPT-5.2 Progpt-5.2-pro | $21.00 | — | — | $168.00 | — | — | $0.11 |
| GPT-5.1 and 5 | |||||||
| GPT-5.1gpt-5.1 | $1.25 | $0.125 | — | $10.00 | — | — | $0.0063 |
| GPT-5gpt-5 | $1.25 | $0.125 | — | $10.00 | — | — | $0.0063 |
| GPT-5 Progpt-5-pro | $15.00 | — | — | $120.00 | — | — | $0.075 |
| GPT-5 minigpt-5-mini | $0.25 | $0.025 | — | $2.00 | — | — | $0.0013 |
| GPT-5 nanogpt-5-nano | $0.05 | $0.005 | — | $0.40 | — | — | $0.00025 |
| GPT-4.1 and GPT-4o | |||||||
| GPT-4.1gpt-4.1 | $2.00 | $0.50 | — | $8.00 | — | — | $0.006 |
| GPT-4.1 minigpt-4.1-mini | $0.40 | $0.10 | — | $1.60 | — | — | $0.0012 |
| GPT-4.1 nanogpt-4.1-nano | $0.10 | $0.025 | — | $0.40 | — | — | $0.0003 |
| GPT-4ogpt-4o | $2.50 | $1.25 | — | $10.00 | — | — | $0.0075 |
| GPT-4o minigpt-4o-mini | $0.15 | $0.075 | — | $0.60 | — | — | $0.00045 |
| o-series reasoning | |||||||
| o3o3 | $2.00 | $0.50 | — | $8.00 | — | — | $0.006 |
| o3-proo3-pro | $20.00 | — | — | $80.00 | — | — | $0.06 |
| o3-minio3-mini | $1.10 | $0.55 | — | $4.40 | — | — | $0.0033 |
| o4-minio4-mini | $1.10 | $0.275 | — | $4.40 | — | — | $0.0033 |
| o1o1 | $15.00 | $7.50 | — | $60.00 | — | — | $0.045 |
| o1-proo1-pro | $150.00 | — | — | $600.00 | — | — | $0.45 |
Long-context prices
Rates that apply once a request passes the threshold, per 1M tokens.
| Model | Applies above | Input | Cached input | Output |
|---|---|---|---|---|
| GPT-6 Astra | 272K | $20.00 | $2.00 | $75.00 |
| GPT-6.1 Sol | 272K | $4.00 | $0.20 | $15.00 |
| GPT-6 Sol | 272K | $4.00 | $0.40 | $15.00 |
| GPT-6 Luna | 272K | $0.20 | $0.02 | $0.75 |
| GPT-5.6 Sol | 272K | $8.00 | $0.80 | $30.00 |
| GPT-5.6 Terra | 272K | $4.00 | $0.40 | $18.00 |
| GPT-5.6 Luna | 272K | $0.40 | $0.04 | $1.80 |
| GPT-5.5 | 272K | $10.00 | $1.00 | $45.00 |
| GPT-5.4 | 272K | $5.00 | $0.50 | $22.50 |
When promotional prices end
These models are on launch or promotional pricing. The table shows the current rate next to the rate OpenAI has announced for afterwards.
| Model | Current price until | Input now → later | Output now → later |
|---|---|---|---|
| GPT-5.6 Sol | at least Nov 21, 2026 | $4.00 → not announced | $20.00 → not announced |
Context windows
| GPT-6 Astra | 1.05M tokens |
|---|---|
| GPT-6.1 Sol | 1.05M tokens |
| GPT-6 Sol | 1.05M tokens |
| GPT-6 Luna | 1.05M tokens |
Recent OpenAI changes
Model launches, price changes and retirements, newest first.
gpt-4o and gpt-4.1 snapshots deprecated, shutdown January 14, 2027
Replacement: gpt-6-luna.
vendorOpenAI deprecations
GPT-5.5 leaves ChatGPT and Codex
Replaced by GPT-6.1 Sol on paid plans and GPT-6 Luna on Free and Go. The API is not affected.
vendorOpenAI Codex models
Ultrafast processing for GPT-6.1 Sol
6x the standard price: $12 / $60 per 1M tokens.
vendorOpenAI API changelog
GPT-6 becomes the default model in ChatGPT
GPT-6 Sol on paid plans, GPT-6 Luna on Free and Go (from October 8). The Pro reasoning option uses GPT-6 Astra.
vendorOpenAI
gpt-5.3-codex, gpt-5.1 and gpt-5.4-nano deprecated, shutdown April 1, 2027
Replacements: gpt-6-sol for gpt-5.3-codex and gpt-5.1, gpt-6-luna for gpt-5.4-nano.
vendorOpenAI deprecations
Billing rules that change the price
- GPT-6 is OpenAI's current family: Astra (flagship, September 3, 2026), Sol and Luna (September 22) and GPT-6.1 Sol (September 29), the newer Sol. OpenAI calls GPT-6 Sol and Luna direct successors to GPT-5.6 Sol and Luna at half the price; the GPT-5.6 models are not deprecated in the API.
- GPT-5.6 Sol's $4 / $20 price is promotional and available at least through November 21, 2026. OpenAI cut it on August 21, 2026, by 20% on input and 33% on output, and has not said what it will cost after the promotion.
- Batch API: 50% off both input and output. Flex processing is priced like Batch on select models.
- Fast mode costs 2x the standard rate. It is the new name for Priority processing since July 30, 2026, and requests tagged priority still work.
- Cache writes are listed for the GPT-6 and GPT-5.6 families at 1.25x the input price (2.5x above 272K tokens).
- Regional processing (data residency) endpoints add 10% for eligible models released on or after March 5, 2026.
- The GPT-6 family, the GPT-5.6 family, GPT-5.5 and GPT-5.4 switch to long-context rates once a prompt exceeds 272K input tokens.
- Pro models have no cached-input discount.
- The gpt-5.6 alias routes requests to gpt-5.6-sol.
- Ultrafast processing costs 6x the standard rate and is offered only for GPT-6 Astra (US data residency only) and GPT-6.1 Sol.
- GPT-6.1 Sol's cached input is 95% off the input price; the other GPT-6 models give 90% off.
Questions
How much does the OpenAI API cost?
OpenAI lists 32 models. Input costs from $0.05 (GPT-5 nano) to $150.00 (o1-pro) per 1M tokens, and output from $0.40 (GPT-5 nano) to $600.00 (o1-pro). A request with 1,000 input and 500 output tokens costs between $0.00025 and $0.45.
What is the cheapest OpenAI API model?
By the cost of a 1,000-input, 500-output request, GPT-5 nano is the cheapest: $0.05 input and $0.40 output per 1M tokens, about $0.00025 per request.
Does the OpenAI API have a batch discount?
Yes. The Batch API costs 50% less than standard rates, for both input and output.
Related
Where these numbers come from
Prices are copied from the vendor's own pages; the vendor page is the authority on what you are billed. Per-request costs are our arithmetic.
| Field | Source | Observed |
|---|---|---|
| API prices per 1M tokens, all models | vendorOpenAI API pricing | Oct 17, 2026 |
| Shutdown schedule | vendorOpenAI deprecations | Oct 17, 2026 |
| Cost of 1,000 input + 500 output tokens | derivedCalculated from the prices above | Oct 17, 2026 |