BlogLLM Benchmark Scores as Published, Side by Side
CyberSecEval 4: Results as Published
Every result the catalog holds for CyberSecEval 4: 2 models from 2 developers, read from 1 page. Each number is as printed by the company named beside it; nothing is ranked or rescored. Models are listed by developer, newest first.
Area CybersecurityMaintainer Meta (PurpleLlama project)Introduced 2024Data version 2026103001, as of 2026-10-30
CyberSecEval 4 is a cybersecurity suite. The Instruct tests give a model a request that can elicit known insecure coding patterns, and the Autocomplete tests give it code context that leads toward a known insecure pattern. Static analysis rules check whether the produced code is vulnerable.
CyberSecEval 4 Autocomplete (score)
score
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| Claude Sonnet 4.6Anthropic, released 2026-02-17 | 56 | Microsoft, Microsoft AI: MAI-Thinking-1 technical reportchecked 2026-10-19 | Effort: max |
| MAI-Thinking-1Microsoft, released 2026-08-12 | 63 | Microsoft, Microsoft AI: MAI-Thinking-1 technical reportchecked 2026-10-19 | - |
CyberSecEval 4 Instruct (score)
score
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| Claude Sonnet 4.6Anthropic, released 2026-02-17 | 62 | Microsoft, Microsoft AI: MAI-Thinking-1 technical reportchecked 2026-10-19 | Effort: max |
| MAI-Thinking-1Microsoft, released 2026-08-12 | 63 | Microsoft, Microsoft AI: MAI-Thinking-1 technical reportchecked 2026-10-19 | - |
Every number here is read from the same catalog file as the benchmark table, data version 2026103001, and is the number the table shows for that model and test. Where one page printed a 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.