How much does GPT-5.5 cost per million tokens?
GPT-5.5 costs $5.00 per 1M input tokens and $30.00 per 1M output tokens.
GPT-5.5 API Pricing
Verified standard token pricing for GPT-5.5, with official-source evidence and computed workload costs.
As of Sep 20, 2026, GPT-5.5 costs $5.00 per 1M input and $30.00 per 1M output tokens, $0.50 cached input. At standard non-cached rates, 10,000 monthly requests using 800 input and 300 output tokens each cost about $130.00, before retries, tools, storage, or provider-specific discounts.
| Model | Input $/1M | Cached input $/1M | Output $/1M | Source | Last verified |
|---|---|---|---|---|---|
| GPT-5.5 | $5.00 | $0.50 | $30.00 | Official source | Sep 20, 2026 |
Prices from official provider pages, verified on the date shown.
Official source detail
Full pricing details
Standard prices are shown per 1M tokens. Provider-specific tiers, components, Batch rates, and time schedules remain separate below.
Long-context pricing tiers
| Tier | Input | Cached input | Output |
|---|---|---|---|
| Long context | $10.00 | $1.00 | $45.00 |
| Short context | $5.00 | $0.50 | $30.00 |
Published pricing components
| Mode and context | Component | Modality | Price |
|---|---|---|---|
| batch · long | cached input | text | $0.5 / per 1m tokens |
| batch · long | input | text | $5 / per 1m tokens |
| batch · long | output | text | $22.5 / per 1m tokens |
| batch · short | cached input | text | $0.25 / per 1m tokens |
| batch · short | input | text | $2.5 / per 1m tokens |
| batch · short | output | text | $15 / per 1m tokens |
| flex · long | cached input | text | $0.5 / per 1m tokens |
| flex · long | input | text | $5 / per 1m tokens |
| flex · long | output | text | $22.5 / per 1m tokens |
| flex · short | cached input | text | $0.25 / per 1m tokens |
| flex · short | input | text | $2.5 / per 1m tokens |
| flex · short | output | text | $15 / per 1m tokens |
| standard · long | cached input | text | $1 / per 1m tokens |
| standard · long | input | text | $10 / per 1m tokens |
| standard · long | output | text | $45 / per 1m tokens |
| standard · short | cached input | text | $0.5 / per 1m tokens |
| standard · short | input | text | $5 / per 1m tokens |
| standard · short | output | text | $30 / per 1m tokens |
Batch API: input $2.50 and output $15.00 per 1M tokens.
Official pricing notes: Long-context standard pricing is higher: $10 input, $1 cached input, $45 output per 1M tokens.
Computed examples
What it costs for real workloads
| Workload | Monthly volume | Input tokens | Output tokens | Estimated cost |
|---|---|---|---|---|
| Side project | 1,000 requests/month | 800,000 | 300,000 | $13.00 |
| Side project | 10,000 requests/month | 8,000,000 | 3,000,000 | $130.00 |
| Side project | 100,000 requests/month | 80,000,000 | 30,000,000 | $1,300.00 |
| Customer support chatbot | 10,000 conversations/month | 121,650,000 | 10,800,000 | $932.25 |
| Agent task | 1,000 tasks/month | 12,000,000 | 3,200,000 | $156.00 |
| Side project with Batch API | 10,000 requests/month | 8,000,000 | 3,000,000 | $65.00 |
Assumptions
- Side project: 800 input and 300 output tokens per request at 1,000, 10,000, and 100,000 requests per month.
- Customer support chatbot: 6 replies per conversation, 600 system tokens, 90 user-message tokens, 1,000 retrieved-context tokens, 180 output tokens per reply, and 50% history carry, for 10,000 conversations.
- Agent task: 8 model calls per task, 1,500 input and 400 output tokens per call, for 1,000 tasks.
- Standard rows use non-cached, non-batch prices. A Batch row appears only when both published Batch token prices exist.
Availability evidence
Lifecycle
- Status
- active
- Release date
- Apr 24, 2026
- Context window
- Short / long
- Pricing basis
- V2 projection default price
Nearby blended prices
Compare with
Alternatives are nearest by a 3:1 input-to-output blended token price, with same-provider coverage when available.
| Model | Input / 1M | Output / 1M | Side project, 10k requests |
|---|---|---|---|
| GPT-5.6 Sol | $4.00 | $20.00 | $92.00 |
| Claude Opus 4.8 | $5.00 | $25.00 | $115.00 |
| Claude Opus 5 | $5.00 | $25.00 | $115.00 |
Next steps
Related
Common questions
Frequently asked questions
Does GPT-5.5 have cached input pricing?
Yes. The published cached-input price is $0.50 per 1M tokens.
How much does GPT-5.5 cost for 10,000 requests a month?
Using 800 input and 300 output tokens per request at standard non-cached rates, 10,000 monthly requests cost about $130.00.
Where does this GPT-5.5 price come from?
The prices come from OpenAI's official pricing source and were verified on Sep 20, 2026. See the pricing dataset methodology.