ChatGPT plans and pricing for every use case
ChatGPT pricing, broken down by plan
Everyday AI assistant for chatting, coding, writing, and brainstorming
Great At
- Answering questions and explanations
- Writing and debugging code
- Creative writing and brainstorming
- Data analysis with file uploads and image generation
Cost watchouts
- Free tier access is limited
- Subscription charges do not include API usage
- Pro plans run $100 to $500 per month
Best all-rounder AI. Get Plus ($20/mo) if you use it daily.
| Plan | Price | What You Get | Best For |
|---|---|---|---|
| Free | $0 | Free text chats with GPT-6 Luna, subject to abuse guardrails; uploads, image creation, voice, deep research, memory and Codex access are limited. | Trying ChatGPT with limited usage |
| Go | $8/mo | More messages with tools, uploads, image creation and voice chats than Free, plus longer memory. This plan may include ads. | Longer conversations, more uploads and everyday tool use |
| Plus | $20/mo | Advanced reasoning, expanded messages, uploads and deep research; projects, scheduled tasks, custom GPTs and expanded Codex access. Limits apply. | Advanced work with research, projects and scheduled tasks |
| Pro 100 | $100/mo | $100/month. Pro models, Codex and deep research; no Ultrafast. Usage allowances vary by model. | Codex, research and Pro models with a lower monthly budget |
| Pro 200 | $200/mo | $200/month. New subscriptions have reopened. More included usage than Pro 100, with a lower allowance than previously offered; no Ultrafast. Eligible existing subscribers keep the previous allowance through October 29, 2026 while subscribed. | Codex, research and Pro models with more included usage |
| Business Standard | $25/user/mo ($20/user/mo annual)Annual option: $20/month equivalent | $25 per user with monthly billing or $20 per user per month with annual billing. Business workspaces require at least two seats. | Teams that need a shared ChatGPT workspace |
| Business Premium | $125/user/mo ($100/user/mo annual)Annual option: $100/month equivalent | $125 per user with monthly billing or $100 per user per month with annual billing. Premium includes five times more usage than Standard; Business workspaces require at least two seats. | Teams that need higher Business usage limits |
| Enterprise | Contact sales | Pricing is not publicly listed. | Organizations needing custom governance and terms |
| Pro 500 | $500/month | Highest included usage among the three Pro plans; includes Ultrafast. Usage allowances vary by model. | Heavy Pro usage requiring Ultrafast |
Review provenance
Check source · Check source · Check source · Check source · Check source · Check source · Check sourceFree plan and entry cost
A public free plan is available. Model access and usage limits can change, so check the official plan details before relying on a specific model.
How to read the total cost
- Billing and usageGPT-6.1 Sol standard short-context API rates are $2 input, $0.10 cached input, $2.50 cache write, and $10 output per 1 million tokens. GPT-6 Luna rates are $0.10, $0.01, $0.125, and $0.50. Standard global short-context rates apply up to 272,000 input tokens. Above that threshold, the full request uses higher rates. GPT-6.1 Sol long-context rates are $4 input, $0.20 cache read, $5 cache write and $15 output per million tokens. Optional Astra Ultrafast global short-context rates are $60 input, $6 cache read, $75 cache write and $300 output; long-context rates are $120, $12, $150 and $450. Standard Astra short-context rates are $10, $1, $12.50 and $50. Optional 12-hour cache writes (added October 20, 2026) cost 2.5x input: $5 for Sol, $25 for Astra and $0.25 for Luna per million short-context tokens. Optional GPT-6.1 Sol Ultrafast global short-context rates are $12 input, $0.60 cache read, $15 cache write and $60 output; long-context rates are $24, $1.20, $30 and $90. Ultrafast supports global processing and US data residency only, subject to rate limits. API rates do not establish ChatGPT subscription allowance changes.
- Pro allowancesPro 100, Pro 200 and Pro 500 replace the previous multiplier labels. Grandfathering applies to an active Pro 200 subscription at any point September 22 through September 29, 2026 at 10 a.m. Pacific Time, and lasts through October 29 while the subscription stays active. It does not grant Pro 500 features.
- Review date scopeThe displayed record review date is the earliest retained review, not the date every claim was last checked. Pro prices, availability, Ultrafast eligibility and grandfathering were checked against the retained October 7 official help extraction. The API ratecard retains October 5 evidence; other subscription fields retain their earlier sources. The separate October 7 browser capture was blocked.
Price History
No independently verified historical ChatGPT subscription price series is supplied in this audit. Current Pro tier changes are evidenced by the October 4 help-page snapshot, not manufactured chart points. API pricing updates are separate from subscription prices.
Price Trend by Plan
No independently verified multi-date ChatGPT subscription price series is available.
Pricing Calculator
More messages with tools, uploads, image creation and voice chats than Free, plus longer memory. This plan may include ads.
Extra monthly cost
Add a total for the whole stack, not a per-seat price. Use your own estimate or quote; this cost is not verified.
Projected from monthly payments, not an annual invoice.
Annual and quarterly billing show a monthly average. Extra costs and custom quotes are your estimate.
How this estimate works
Extra costs are your total estimate for the whole stack. Per-unit cost divides that total across the selected quantity; dedicated member and add-on pricing is not modelled.
Subscription vs API
ChatGPT subscriptions and the API platform use separate billing systems. API usage is billed independently from ChatGPT subscription charges.
| Model | Input | Cache read | Cache write | Output |
|---|---|---|---|---|
| GPT-6.1 Sol (standard, short context) | $2 | $0.1 | $2.5 | $10 |
| GPT-6 Luna (standard, short context) | $0.1 | $0.01 | $0.125 | $0.5 |
| GPT-6.1 Sol (standard, long context) | $4 | $0.2 | $5 | $15 |
| GPT-6 Sol (standard, short context) | $2 | $0.2 | $2.5 | $10 |
| GPT-6 Astra (standard, short context) | $10 | $1 | $12.5 | $50 |
| GPT-6 Astra (standard, long context) | $20 | $2 | $25 | $75 |
| GPT-6.1 Sol (ultrafast, short context) | $12 | $0.6 | $15 | $60 |
| GPT-6.1 Sol (ultrafast, long context) | $24 | $1.2 | $30 | $90 |
| GPT-6 Astra (ultrafast, short context) | $60 | $6 | $75 | $300 |
| GPT-6 Astra (ultrafast, long context) | $120 | $12 | $150 | $450 |
USD per million tokens, global processing. Short context is up to 272,000 input tokens; above 272,000 input tokens the full request uses long-context rates. GPT-6.1 Sol is a new model with a 50% lower cache-read rate than GPT-6 Sol, not a retroactive repricing or 50% lower total bill. GPT-6 Astra Ultrafast is optional in the Responses API and subject to rate limits. It supports global processing and US data residency only; EU and other regional inference residency are unsupported. GPT-6.1 Sol Ultrafast (added October 8, 2026) is also optional and rate-limited, with global processing and US and EU data residency. Regional processing adds a separate uplift. API prices are separate from ChatGPT subscription allowances. Since October 20, 2026, Responses API requests can choose 12-hour prompt caching: 12-hour cache writes cost 2.5x input (GPT-6.1 Sol $5 short context, $10 long; GPT-6 Astra $25 and $50; GPT-6 Luna $0.25 and $0.50 per million tokens), against 1.25x for the default 30-minute write. Cache reads cost the same for both.
Pros & Cons
Pros
- Free entry tier and monthly paid plans
- Pro tiers include Codex and deep research
- Pro 500 includes Ultrafast
Cons
- Subscription charges do not include API usage
- Included usage depends on the model
- Pro 200 grandfathering ends October 29, 2026 for eligible subscribers