BlogLLM Benchmark Scores as Published, Side by Side
gpt-5.6-cyber: Benchmark Scores as Published
Every result the catalog holds for gpt-5.6-cyber: 5 tests in 1 area, from 2 pages. All were printed by OpenAI itself. Each number is as printed; nothing is ranked or rescored.
Developer OpenAIContext window 400KMax output 128KInput Text, imageOutput TextWeights ClosedData version 2026103001, as of 2026-10-30
Cybersecurity
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| Advanced Cybersecurity Completion Rate | 95% | OpenAI, GPT-6 Astra System Cardchecked 2026-10-25 | - |
| Cyber red-teaming | 100% | OpenAI, GPT-6 Astra System Cardchecked 2026-10-25 | - |
| Proof-of-concept exploit creation | 98% | OpenAI, GPT-6 Astra System Cardchecked 2026-10-25 | - |
| Vulnerability discovery and analysis | 100% | OpenAI, GPT-6 Astra System Cardchecked 2026-10-25 | - |
| Vulnerability patching | 98% | OpenAI, GPT-6 Astra System Cardchecked 2026-10-25 | - |
Price
API list prices in US dollars per one million tokens, as printed on the developer's pricing page on the day checked.
| Price | Value | Printed by | Setup as printed |
|---|---|---|---|
| Input | $12.5 | OpenAI, OpenAI API pricingchecked 2026-10-15 | - |
| Cached input | $1.25 | OpenAI, OpenAI API pricingchecked 2026-10-15 | - |
| Cache write | $15.625 | OpenAI, OpenAI API pricingchecked 2026-10-15 | - |
| Output | $75 | OpenAI, OpenAI API pricingchecked 2026-10-15 | - |
More from OpenAI
- GPT-6.1 Terra
- GPT-6.1 Sol
- GPT-6 Luna
- GPT-6 Sol
- GPT-6 Astra
- GPT-5.6 Luna
- GPT-5.6 Sol
- GPT-5.6 Terra
- GPT-5.5
- GPT-5.5 Pro
- GPT-5.4 mini
- GPT-5.4 nano
- GPT-5.4
- GPT-5.4 Pro
- GPT-5.3-Codex
- GPT-5.2-Codex
- GPT-5.2
- GPT-5.2 Instant
- GPT-5.2 Pro
- GPT-5.1-Codex-Max
- GPT-5.1
- GPT-5.1 Instant
- GPT-5-Codex
- GPT-5
- GPT-5 mini
- GPT-5 Pro
- gpt-oss-120b
- gpt-oss-20b
- ChatGPT agent
- codex-1
- o3
- OpenAI o4-mini
- GPT-4.1
- GPT-4.5
- Deep research
- o3-mini
- o1
- o1-mini
- GPT-4o
- GPT-5.5 Instant
Every number here is read from the same catalog file as the benchmark table, data version 2026103001. Where one page printed the model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.