OpenAI API Pricing (October 2026): GPT-6, GPT-5.6, Batch, Flex, Fast and Credits
On this page
Short answer: as of October 15, 2026, OpenAI's flagship API models cost $10 / $50 per million input / output tokens (GPT-6 Astra), $2 / $10 (GPT-6.1 Sol) and $0.10 / $0.50 (GPT-6 Luna). Cached input is 90–95% cheaper, the Batch API and Flex processing halve the price, and Fast mode roughly doubles it.
Prices are copied from OpenAI's pricing page on October 15, 2026, in USD per million tokens. "Short context" means prompts up to 272K input tokens. Longer prompts use the long-context rates.
Current flagship models
| Model | Input | Cached input | Cache write | Output | Input (> 272K) | Output (> 272K) |
|---|---|---|---|---|---|---|
| gpt-6-astra | $10.00 | $1.00 | $12.50 | $50.00 | $20.00 | $75.00 |
| gpt-6.1-sol | $2.00 | $0.10 | $2.50 | $10.00 | $4.00 | $15.00 |
| gpt-6-sol | $2.00 | $0.20 | $2.50 | $10.00 | $4.00 | $15.00 |
| gpt-6-luna | $0.10 | $0.01 | $0.125 | $0.50 | $0.20 | $0.75 |
| gpt-5.6-sol | $4.00 | $0.40 | $5.00 | $20.00 | $8.00 | $30.00 |
| gpt-5.6-terra | $2.00 | $0.20 | $2.50 | $12.00 | $4.00 | $18.00 |
| gpt-5.6-luna | $0.20 | $0.02 | $0.25 | $1.20 | $0.40 | $1.80 |
A few things stand out:
- GPT-6.1 Sol is the sweet spot. It costs the same as GPT-6 Sol, its cached input is half the price, and it's the model OpenAI recommends for Codex.
- GPT-5.6 Sol costs twice as much as GPT-6.1 Sol. OpenAI notes promotional pricing on it through at least November 21, 2026. If you're still on it, moving to 6.1 Sol likely cuts your bill.
- Long prompts cost more. Above 272K input tokens, input doubles and output rises 50%.
Older models still available
| Model | Input | Cached input | Output |
|---|---|---|---|
| gpt-5.5 (< 272K) | $5.00 | $0.50 | $30.00 |
| gpt-5.4 (< 272K) | $2.50 | $0.25 | $15.00 |
| gpt-5.4-mini | $0.75 | $0.075 | $4.50 |
| gpt-5.4-nano | $0.20 | $0.02 | $1.25 |
| gpt-5 / gpt-5.1 | $1.25 | $0.125 | $10.00 |
| gpt-5-mini | $0.25 | $0.025 | $2.00 |
| gpt-5-nano | $0.05 | $0.005 | $0.40 |
| gpt-4.1 | $2.00 | $0.50 | $8.00 |
| gpt-4o | $2.50 | $1.25 | $10.00 |
| gpt-4o-mini | $0.15 | $0.075 | $0.60 |
| o3 | $2.00 | $0.50 | $8.00 |
| o4-mini | $1.10 | $0.275 | $4.40 |
Pro variants (gpt-5.5-pro, gpt-5.4-pro) cost $30 / $180, and o1-pro costs $150 / $600. Use them only when you've measured that you need them.
Processing modes: Batch, Flex, Fast
The same model can be billed at different rates depending on how quickly you need the answer:
| Mode | Price vs Standard | How it works | Use it for |
|---|---|---|---|
| Batch | 50% | Submit a file of requests and collect results later | Offline jobs: evaluations, bulk summarization, data labeling |
| Flex | 50% | Normal API calls at lower priority, so responses can be slower | Background work you still want through the regular API |
| Standard | 100% | Normal API calls | Interactive apps |
| Fast | about 2x | Higher-priority, faster processing. Called "Priority" until July 30, 2026 | Latency-sensitive production traffic |
| Ultrafast | 6x (gpt-6-astra and gpt-6.1-sol only) | Fastest tier | Real-time products where speed beats cost |
For example, GPT-6.1 Sol costs $1 / $5 on Batch or Flex, $4 / $20 on Fast and $12 / $60 on Ultrafast.
Other charges
| Item | Price |
|---|---|
| Web search tool | $10 per 1,000 calls, plus search content tokens at the model's rate |
| File search tool | $2.50 per 1,000 calls (Responses API) |
| File search storage | $0.10 per GB per day, with the first GB free |
| Regional processing (data residency) | +10% for models released on or after March 5, 2026 |
The Responses, Chat Completions, Realtime, Batch and Assistants APIs aren't priced separately. You pay the model's token rates.
Credits and usage tiers
The API runs on prepaid credits bought in your OpenAI Platform account. Your organization's usage tier sets both its rate limits and a monthly usage cap, and it rises automatically as your total credit purchases grow:
| Tier | Qualification | Monthly usage limit |
|---|---|---|
| Free | Account in a supported country | $100 |
| Build | $5 in total credit purchases | $500 |
| Launch | $100 in total credit purchases | $5,000 |
| Grow | $500 in total credit purchases | $200,000 |
You can see your tier and per-model rate limits under Settings → Organization → Limits.
ChatGPT plans are separate. A ChatGPT Plus or Pro subscription doesn't add API credits. It covers ChatGPT, plus Codex in the CLI and IDE. If you sign Codex in with an API key instead, it bills at the API prices on this page.
Cost calculator
An estimate at October 2026 list prices, for short-context requests (≤ 272K input tokens), excluding tools and cache-write premiums.
Worked example
A support assistant on gpt-6.1-sol handles 30,000 requests a month. Each request has 3,000 input tokens, 2,000 of them a repeated system prompt served from cache, plus 400 output tokens:
- Uncached input: 30M tokens × $2 = $60
- Cached input: 60M tokens × $0.10 = $6
- Output: 12M tokens × $10 = $120
- Total ≈ $186 a month, or about $93 on Flex if slower responses are fine.
The same workload costs the same on Claude Sonnet 5.5, which has identical $2 / $10 / $0.10 rates. See Claude API pricing to compare model by model.
How to spend less
- Default to GPT-6.1 Sol or GPT-6 Luna, and step up to Astra only for tasks that measurably need it.
- Keep prompts under 272K tokens when you can, because the long-context rates are significantly higher.
- Put stable content first (instructions, tool definitions, reference text) so it's served from cache.
- Use Batch or Flex for anything not user-facing.
- Cap output length. Output costs 5x input on GPT-6.1 Sol and GPT-6 Luna.
Related
FAQ
How much does the OpenAI API cost?
As of October 2026, GPT-6.1 Sol costs $2 per million input tokens and $10 per million output tokens, GPT-6 Astra costs $10 and $50, and GPT-6 Luna costs $0.10 and $0.50. For prompts longer than 272K tokens, input prices double and output prices rise 50% on models with long-context pricing.
What is the cheapest OpenAI model?
Among current models, GPT-6 Luna at $0.10 per million input tokens and $0.50 per million output tokens, and half that through the Batch API or Flex processing. Older models such as gpt-5-nano are cheaper still but less capable.
How do OpenAI API credits work?
The API is billed from prepaid credits. Your organization's usage tier, and its monthly usage limit, rises automatically as total credit purchases reach $5 (Build), $100 (Launch) and $500 (Grow). Free-tier accounts in supported countries have a $100 monthly limit.
What is OpenAI Flex processing?
Flex processing runs requests at lower priority for the same 50% discount as the Batch API, but through the normal synchronous API. It suits background jobs that can tolerate slower responses.
Is ChatGPT Plus the same as API credits?
No. ChatGPT subscriptions and the API are billed separately. A ChatGPT plan includes Codex usage in the CLI and IDE, but API calls from your own code are billed to your OpenAI Platform account at API prices.
Found a mistake? Email [email protected].