- OpenAI Dots price: Your first Dot is included with ChatGPT Pro and Business Premium at no extra charge
- Chat is free: Messaging your Dot does not consume your regular ChatGPT usage limits
- Real work costs: Actual task execution runs through Codex and draws on your Codex usage allowance
- Pro 500 plan: Ultrafast speed is exclusive to the new $500/month Pro 500 tier
- Availability: Pro is adults only and not yet available in Europe, the UK, or Switzerland
OpenAI Dots Price Overview
OpenAI Dots price structure is simpler than most people expect: the Dot itself is a bundled feature, not a standalone subscription. Announced at DevDay 2026, Dots are persistent AI agents that each run on their own cloud computer, connect to more than 4,000 apps, and keep working around the clock. Instead of paying per Dot, you pay for the plan that unlocks them — and the real cost lives in the underlying compute your Dot consumes.
Video Highlights:
- Dots are live on ChatGPT Pro and Business Premium; Enterprise is in beta
- Your first Dot is included, and chatting with it does not touch your usage limits
- Real agent work runs through Codex, which does consume usage
- OpenAI reopened the $200 Pro plan with half the previous Codex allowance
- Early-access testers report Dots are still buggy — factor that into your upgrade decision
The $200 Pro plan was reopened with half the Codex usage it previously included. According to Engadget, current subscribers keep their old limits for roughly one month before the reduction applies. Check your usage before your limits change.
Plan-by-Plan Cost Comparison
Each plan tier unlocks Dots differently, and the pricing reflects how much agent compute you get rather than the Dot feature itself.
| Plan | Dots Access | Dot Cost | Notes |
|---|---|---|---|
| ChatGPT Plus | Not included | N/A | Can use GPT-6.1 Soul in Codex instead |
| ChatGPT Pro ($200/mo) | Yes | First Dot included | Reopened with half the prior Codex allowance |
| Pro 500 ($500/mo) | Yes | Included | Exclusive access to Ultrafast speed in ChatGPT |
| Business Premium | Yes | First Dot included | Best value for small teams |
| Enterprise | Beta | Custom pricing | Specialist Dots previewed for accounting, legal, support |
Charging your Dot with jobs like monitoring bug reports, testing fixes, or drafting PRs consumes Codex usage, not your chat limits. Budget around your Codex allowance, not your message count.
CardGrid comparison of the three main buyer profiles:
Individual Pro User
- $200/month base plan
- First Dot included
- Codex allowance recently halved
- Adults only, not in EU/UK/Switzerland yet
Power User / Pro 500
- $500/month plan
- Ultrafast mode (up to 8x faster in Codex)
- Heaviest agent workloads
- Best fit for high-volume automation
Business / Team
- Business Premium pricing per seat
- Shared ChatGPT Spaces included
- Specialist Dots for team workflows
- Shared slides coming in following weeks
Underlying Model Pricing: Astra vs Soul
Dots and Codex run on OpenAI's models, and model choice directly affects your effective cost per task. GPT-6.1 Soul was launched at DevDay 2026 as the budget-friendly option.
| Model | Input Price (per 1M tokens) | Output Price (per 1M tokens) | Performance Notes |
|---|---|---|---|
| GPT-6 Astra | $10 | $50 | Flagship model |
| GPT-6.1 Soul | $2 | $10 | Ties Astra on the deep SWEE coding benchmark |
| Ultrafast API mode | 6x normal rate | 6x normal rate | Up to 8x faster execution |
Soul delivers roughly 80% savings versus Astra at one-fifth of the price, while matching Astra's score on OpenAI's internal coding test. It is available in Codex, ChatGPT work, and the API — though not yet in regular ChatGPT chat.
If you are on Plus, GPT-6.1 Soul in Codex is described as the biggest upgrade within a plan you already pay for — no Dots subscription required to benefit from the new pricing tier.
Step-by-Step: Estimating Your Dots Cost
Follow this process to figure out what Dots will actually cost you each month before committing to a plan.
Audit Your Current Plan
Check whether you are on Plus, Pro, or Business Premium. Only Pro, Pro 500, Business Premium, and Enterprise (beta) currently unlock Dots.
Estimate Agent Workload
List the jobs you would delegate — inbox monitoring, bug triage, calendar chasing. Real execution counts against your Codex allowance; conversations with your Dot do not.
Pick Your Model Tier
Route routine agent tasks to GPT-6.1 Soul ($2/$10 per million tokens) and reserve Astra for complex work to keep costs low.
Factor In Speed Needs
Only pay the $500 Pro 500 premium if Ultrafast speed genuinely matters — in the API it costs six times the normal rate.
Start Small and Validate
Give your first Dot one small job, such as watching a single inbox. Confirm you trust the output before scaling up workload and cost.
Specialist Dots for accounting, legal review, and support tickets were previewed for larger organizations. Enterprise Dots access is currently in beta with custom pricing.
Pre-Purchase Checklist
Before you pay for a plan to get Dots, run through this checklist:
Before You Buy:
- Confirm Dots are available in your region (not yet in Europe, UK, Switzerland)
- Verify your plan includes a first Dot at no extra charge
- Review your current Codex usage against the halved Pro allowance
- Decide whether Ultrafast speed justifies the Pro 500 premium
- Test GPT-6.1 Soul pricing in Codex before committing to agent workloads
Reviewer Dan Shipper at Every, who had early access, reported Dots are too buggy to recommend at launch — and even the on-stage demo stalled mid-presentation. Price in some patience alongside your subscription.
FAQ
Q: How much does OpenAI Dots cost?
Your first Dot is included with ChatGPT Pro ($200/month), Pro 500 ($500/month), and Business Premium. There is no separate per-Dot charge, but real task execution consumes your Codex usage allowance.
Q: Does chatting with my Dot use my message limits?
No. Conversations with your Dot do not touch your ChatGPT limits. However, actual work the Dot performs runs through Codex, which does count against your usage.
Q: Is there a cheaper way to get Dots-like automation?
ChatGPT Plus users cannot access Dots yet, but GPT-6.1 Soul is available in Codex at $2 per million input tokens and $10 per million output tokens — about one-fifth of Astra's pricing.
Q: What changed with the $200 Pro plan?
OpenAI reopened the $200 Pro plan with half the Codex usage it previously included. Existing subscribers retain their old limits for roughly one month before the reduction takes effect.