GPT 5.5 pricing: real cost per million tokens
What OpenAI's GPT 5.5 really costs across 3 plans from 1 provider: monthly fees, usable allowances and real per-token prices, against the official API list price and independent leaderboards.
Data updated · How real prices are computed
As of 2026-10-08, the cheapest way to use GPT 5.5 is ChatGPT Pro 20x at a real cost of about $0.02004 per million tokens ($200/month for roughly 9.982B tokens a month). We track 3 plans and API channels that serve GPT 5.5. Official API list price per million tokens: $0.5 cached input, $5 input, $30 output — about $0.76 per million tokens on a standard workload mix.
- Lowest real price
- $0.02004/MTokChatGPT Pro 20x
- API list price
- $0.76/MTokstandard workload mix
- Most monthly tokens
- 9.982BChatGPT Pro 20x
- Plans & APIs
- 33 subscriptions · 0 metered APIs
Plot GPT 5.5 on the price–capability chart →
GPT 5.5 plans and real prices
Real price = monthly fee ÷ tokens you can actually use in a month. “vs API” shows how many times cheaper than the official API list price (standard workload mix) the plan is. “Evidence” opens the full calculation and sources for that row.
| Plan | Provider | Billing | Monthly fee | Monthly tokens | Real $/MTok | vs API | Confidence | Data date | |
|---|---|---|---|---|---|---|---|---|---|
| ChatGPT Pro 20x | OpenAI | Subscription | $200/mo | 9.982B | $0.02004 | 37.9× | Medium | Evidence ↓ | |
| ChatGPT Plus | OpenAI | Subscription | $20/mo | 0.477B | $0.04196 | 18.1× | Medium | Evidence ↓ | |
| ChatGPT Pro 5x | OpenAI | Subscription | $100/mo | 2.383B | $0.04197 | 18.1× | Medium | Evidence ↓ |
Behind each number
For every plan: the data date, the exact calculation, the quota basis, how leaderboard scores were mapped, and the original adoption evidence with its sources.
ChatGPT Pro 20xOpenAI$0.0200359/MTok
- Plan
- ChatGPT Pro 20x · OpenAI
- Data date
- 2026-06-10 – 2026-09-06 · derived; uses the data date of ChatGPT Pro 20x · gpt-5.6-sol
- Real price / MTok
- $0.0200359
- Monthly tokens
- 9,982,100,000 (9.982B)
- Quota confidence
- Medium
- Calculation
- $200 / month ÷ 9,982,100,000 tokens × 1,000,000 ≈ $0.0200359 / MTok
- Quota basis
Not workload-normalized: sample lacks a token breakdown (or is an official absolute token table); raw tokens used as-is
Official API list (cached / in / out, per MTok): $0.5 / $5 / $30 · standard-load blended ≈ $0.76/MTok
- Benchmark mapping
- Artificial Analysis Intelligence Index v4.3: Medium (5 configurations)
Terminal-Bench 4.0 (AA): Low (3 configurations)
Code Arena | WebDev Overall: Low (3 configurations)
Agent Arena · Overall (Ranking / Pareto): Low (2 configurations)
DeepSWE v1.1: Low (4 configurations)Mapping confidence: how directly each score refers to the model served here.
Adoption evidence · source and rationale (translated from the Chinese record)
Derived from same-plan gpt-5.6-sol 12.902B × 0.7736842105263158
https://learn.chatgpt.com/docs/pricing three-part credits (cache/input/output) Sol=10/100/500, vs (12.5, 125, 750); old multiplier 0.8, old monthly allowance 10.3216B voided; the Sol baseline is kept, recomputed under the project standard workload; see audit-round4-2026-09-05.json
https://learn.chatgpt.com/docs/pricingaudit-round4-2026-09-05.jsonBrowse the public evidence archive
ChatGPT PlusOpenAI$0.0419639/MTok
- Plan
- ChatGPT Plus · OpenAI
- Data date
- 2026-07-30 · derived; uses the data date of ChatGPT Plus · gpt-5.6-sol
- Real price / MTok
- $0.0419639
- Monthly tokens
- 476,600,000 (0.477B)
- Quota confidence
- Medium
- Calculation
- $20 / month ÷ 476,600,000 tokens × 1,000,000 ≈ $0.0419639 / MTok
- Quota basis
Not workload-normalized: sample lacks a token breakdown (or is an official absolute token table); raw tokens used as-is
Official API list (cached / in / out, per MTok): $0.5 / $5 / $30 · standard-load blended ≈ $0.76/MTok
- Benchmark mapping
- Artificial Analysis Intelligence Index v4.3: Medium (5 configurations)
Terminal-Bench 4.0 (AA): Low (3 configurations)
Code Arena | WebDev Overall: Low (3 configurations)
Agent Arena · Overall (Ranking / Pareto): Low (2 configurations)
DeepSWE v1.1: Low (4 configurations)Mapping confidence: how directly each score refers to the model served here.
Adoption evidence · source and rationale (translated from the Chinese record)
Derived from same-plan gpt-5.6-sol 616M × 0.7736842105263158
https://learn.chatgpt.com/docs/pricing three-part credits (cache/input/output) Sol=10/100/500, vs (12.5, 125, 750); old multiplier 0.8, old monthly allowance 492.8M voided; the Sol baseline is kept, recomputed under the project standard workload; see audit-round4-2026-09-05.json
https://learn.chatgpt.com/docs/pricingaudit-round4-2026-09-05.jsonBrowse the public evidence archive
ChatGPT Pro 5xOpenAI$0.0419657/MTok
- Plan
- ChatGPT Pro 5x · OpenAI
- Data date
- 2026-07-30 · derived; uses the data date of ChatGPT Plus · gpt-5.6-sol
- Real price / MTok
- $0.0419657
- Monthly tokens
- 2,382,900,000 (2.383B)
- Quota confidence
- Medium
- Calculation
- $100 / month ÷ 2,382,900,000 tokens × 1,000,000 ≈ $0.0419657 / MTok
- Quota basis
Not workload-normalized: sample lacks a token breakdown (or is an official absolute token table); raw tokens used as-is
Official API list (cached / in / out, per MTok): $0.5 / $5 / $30 · standard-load blended ≈ $0.76/MTok
- Benchmark mapping
- Artificial Analysis Intelligence Index v4.3: Medium (5 configurations)
Terminal-Bench 4.0 (AA): Low (3 configurations)
Code Arena | WebDev Overall: Low (3 configurations)
Agent Arena · Overall (Ranking / Pareto): Low (2 configurations)
DeepSWE v1.1: Low (4 configurations)Mapping confidence: how directly each score refers to the model served here.
Adoption evidence · source and rationale (translated from the Chinese record)
Derived from same-plan gpt-5.6-sol 3.08B × 0.7736842105263158
https://learn.chatgpt.com/docs/pricing three-part credits (cache/input/output) Sol=10/100/500, vs (12.5, 125, 750); old multiplier 0.8, old monthly allowance 2.464B voided; the Sol baseline is kept, recomputed under the project standard workload; see audit-round4-2026-09-05.json
https://learn.chatgpt.com/docs/pricingaudit-round4-2026-09-05.jsonBrowse the public evidence archive
GPT 5.5 benchmark scores
Highest archived configuration per leaderboard; ranks are among all models each board archives in this dataset. Scores are never mixed across boards.
| Leaderboard | Metric | Score | Rank | Configuration | Effort | Harness | Configs | Snapshot |
|---|---|---|---|---|---|---|---|---|
| Artificial Analysis Intelligence Index v4.3 | Intelligence Index | 38.36 | #27 / 60 | GPT-5.5 (xhigh) | xhigh | — | 5 | 2026-10-01 |
| Terminal-Bench 4.0 (AA) | Resolution Rate % | 14.65 | #24 / 53 | Artificial Analysis - GPT-5.5 (xhigh) | xhigh | Artificial Analysis | 3 | 2026-10-01 |
| Code Arena | WebDev Overall | Arena Score | 1,512 | #37 / 54 | gpt-5.5-xhigh (codex-harness) | xhigh | Codex | 3 | 2026-09-30 |
| Agent Arena · Overall (Ranking / Pareto) | Net Improvement % | 4.12 | #12 / 39 | GPT 5.5 (xHigh) | xhigh | — | 2 | 2026-09-30 |
| DeepSWE v1.1 | Pass@1 % | 67.04 | #15 / 33 | mini-swe-agent - gpt-5-5 (xhigh) | xhigh | mini-swe-agent | 4 | 2026-09-03 |
Every archived configuration
Artificial Analysis Intelligence Index v4.3 5 configurations
| Configuration | Score | Score interval | Harness / effort / mode | Source task cost (mean / median) | Checked | Links |
|---|---|---|---|---|---|---|
| GPT-5.5 (xhigh) | 38.3556 | — | — / xhigh / — | — / — | 2026-10-01 | Source · Archive |
| GPT-5.5 (high) | 36.9794 | — | — / High / — | — / — | 2026-10-01 | Source · Archive |
| GPT-5.5 (medium) | 33.8044 | — | — / Medium / — | — / — | 2026-10-01 | Source · Archive |
| GPT-5.5 (low) [AA estimate]GPT-5.5 (low)Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured. | 30.7095 AA estimate | — | — / Low / — | — / — | 2026-10-01 | Source · Archive |
| GPT-5.5 (non-reasoning) [AA estimate]GPT-5.5 (non-reasoning)Marked intelligenceIndexIsEstimated=true in AA embedded payload; retained as estimate, not treated as fully measured. | 23.1674 AA estimate | — | — / — / — | — / — | 2026-10-01 | Source · Archive |
Terminal-Bench 4.0 (AA) 3 configurations
| Configuration | Score | Score interval | Harness / effort / mode | Source task cost (mean / median) | Checked | Links |
|---|---|---|---|---|---|---|
| Artificial Analysis - GPT-5.5 (xhigh)AA's own-harness independent measurement of Terminal-Bench 4.0 (the terminalBench40 field in the models-board payload, fraction×100); scored separately from the official tbench.ai board. | 14.6465 | — | Artificial Analysis / xhigh / — | — / — | 2026-10-01 | Source · Archive |
| Artificial Analysis - GPT-5.5 (high)AA's own-harness independent measurement of Terminal-Bench 4.0 (the terminalBench40 field in the models-board payload, fraction×100); scored separately from the official tbench.ai board. | 9.0909 | — | Artificial Analysis / High / — | — / — | 2026-10-01 | Source · Archive |
| Artificial Analysis - GPT-5.5 (medium)AA's own-harness independent measurement of Terminal-Bench 4.0 (the terminalBench40 field in the models-board payload, fraction×100); scored separately from the official tbench.ai board. | 5.0505 | — | Artificial Analysis / Medium / — | — / — | 2026-10-01 | Source · Archive |
Code Arena | WebDev Overall 3 configurations
| Configuration | Score | Score interval | Harness / effort / mode | Source task cost (mean / median) | Checked | Links |
|---|---|---|---|---|---|---|
| gpt-5.5-xhigh (codex-harness) | 1,512 | 1,506.1 – 1,517.9 | Codex / xhigh / — | — / — | 2026-10-01 | Source · Archive |
| gpt-5.5-high (codex-harness) | 1,487 | 1,481.1 – 1,492.9 | Codex / High / — | — / — | 2026-10-01 | Source · Archive |
| gpt-5.5 (codex-harness) | 1,455 | 1,449.1 – 1,460.9 | Codex / — / — | — / — | 2026-10-01 | Source · Archive |
Agent Arena · Overall (Ranking / Pareto) 2 configurations
| Configuration | Score | Score interval | Harness / effort / mode | Source task cost (mean / median) | Checked | Links |
|---|---|---|---|---|---|---|
| GPT 5.5 (xHigh) | 4.12 | 3.07 – 5.17 | — / xhigh / — | — / $1.22 | 2026-10-01 | Source · Archive |
| GPT 5.5 | 1.13 | 0.16 – 2.1 | — / — / — | — / $0.61 | 2026-10-01 | Source · Archive |
DeepSWE v1.1 4 configurations
| Configuration | Score | Score interval | Harness / effort / mode | Source task cost (mean / median) | Checked | Links |
|---|---|---|---|---|---|---|
| mini-swe-agent - gpt-5-5 (xhigh) | 67.0354 | 60.57 – 73.501 | mini-swe-agent / xhigh / — | $7.226 / $6.11 | 2026-09-12 | Source · Archive |
| mini-swe-agent - gpt-5-5 (high) | 64.3805 | 61.264 – 67.497 | mini-swe-agent / High / — | $5.1 / $4.583 | 2026-09-12 | Source · Archive |
| mini-swe-agent - gpt-5-5 (medium) | 53.9823 | 51.429 – 56.535 | mini-swe-agent / Medium / — | $2.749 / $2.437 | 2026-09-12 | Source · Archive |
| mini-swe-agent - gpt-5-5 (low) | 26.9912 | 24.697 – 29.286 | mini-swe-agent / Low / — | $1.2 / $1.108 | 2026-09-12 | Source · Archive |
These are reference mappings, not benchmarks of a subscription or API channel. Harness and quota-measurement effort alignment are unverified. Source task costs are not subscription task costs.
- Exact served-model reference; quota-measurement effort is unverified.
- Exact served-model reference only; product harness and quota-measurement effort are unverified. Not a benchmark measurement of this subscription or API channel.
GPT 5.5 official API pricing
| Cached input | Input | Output |
|---|---|---|
| $0.5 | $5 | $30 |
Frequently asked questions
How much does GPT 5.5 really cost per million tokens?
Measured as monthly fee ÷ tokens you can actually use, GPT 5.5 ranges from $0.02004 (ChatGPT Pro 20x) to $0.04197 (ChatGPT Pro 5x) per million tokens.
Which subscription gives the most GPT 5.5 usage?
ChatGPT Pro 20x allows roughly 9.982B tokens of GPT 5.5 per month for $200.
What is the official API price of GPT 5.5?
Official API list price per million tokens: $0.5 cached input, $5 input, $30 output — about $0.76 per million tokens on a standard workload mix.
How does GPT 5.5 score on independent leaderboards?
GPT 5.5 has scores on 5 leaderboards; its strongest placement is 4.12 (Net Improvement %) on Agent Arena · Overall (Ranking / Pareto), #12 of 39 models tracked.
Who makes GPT 5.5, and where can I use it?
GPT 5.5 (model ID gpt-5.5) is made by OpenAI and is available through OpenAI.
More OpenAI models
| Model | Lowest real $/MTok | Cheapest plan | Plans |
|---|---|---|---|
| GPT 5.6 Luna | $0.00142 | ChatGPT Pro 20x | 7 |
| GPT-6 Luna | $0.00198 | Droid Max | 4 |
| GPT 5.6 Terra | $0.00801 | ChatGPT Pro 20x | 5 |
| GPT 5.6 Sol | $0.0155 | ChatGPT Pro 20x | 6 |
| GPT-6 Sol | $0.01818 | ChatGPT Plus | 2 |
| GPT-6.1 Sol | $0.02299 | ChatGPT Plus | 1 |
| GPT-6 Astra | $0.05247 | ChatGPT Pro 20x | 5 |