GPT-5.5 Cyber API Pricing
GPT-5.5 Cyber is the earlier model in OpenAI's Daybreak cybersecurity programme, sold only to approved defenders. OpenAI's price list still carries it at $12.50/M input, $1.25/M cached input and $75.00/M output - the same as GPT-5.6 Cyber, but with no cache-write rate. OpenAI publishes no model page for it. Checked by hand against developers.openai.com.
Run the numbers.
Calculator pre-loaded with GPT-5.5 Cyber's listed rates. With no cache-write rate on the card, a cache miss bills at the input rate - which makes the blended rate slightly lower than GPT-5.6 Cyber's at the same headline prices.
Real-world presets.
Codebase vulnerability sweep
Scanner finding triage
Pentest write-up
Counted with the o200k_base BPE encoding — the same tokenizer the API uses — entirely in your browser. The encoder loads on your first keystroke (~1 MB, one time); your text never leaves this page.
| Model | Input /M | Output /M | Effective blended | Context | Best for |
|---|---|---|---|---|---|
| GPT-5.5 Cyber Current | $12.50 cache $1.25 | $75.00 | $9.01 agentic 92/8 | not listed | The earlier Daybreak cyber model |
| GPT-5.6 Cyber | $12.50 cache $1.25 | $75.00 | $9.53 +6% | 400K | The current Daybreak cyber model |
| GPT-5.5 | $5.00 cache $0.50 | $30.00 | $3.61 −2.5× | 1M | The general GPT-5.5 model |
| Claude Mythos 5.1 | $10.00 cache $0.25 | $50.00 | $6.26 −31% | 1M | Restricted cyber-capable model from another lab |
Frequently asked.
GPT-5.5 Cyber pricing questions - what is published, and what is not.
Q · 01 What is GPT-5.5 Cyber priced at? +
gpt-5.5-cyber at $12.50/M input, $1.25/M cached input and $75.00/M output, with no cache-write rate and no long-context tier.Q · 02 Can I sign up for it? +
Q · 03 Why is so little published about it? +
Q · 04 Is it cheaper than GPT-5.6 Cyber? +
$15.63/M to write a prompt to cache, GPT-5.5 Cyber lists no such rate, so under our standard blend it comes to $9.01/M against $9.53/M.