Benchmark Almanac

CyberOffense

Claude Opus 5.5 87.7% leads since October 2026
Not in model cards

Agents attack the Nova and Lighthouse networks without human help. The score averages the percentage of objectives completed on each range, weighting both equally.

Nova has 21 objectives and Lighthouse has 31. Runs use the Pi agent harness with Kali tools.

Made by
Vals AI
Released
October 2026
Kind
Capture-the-flag challenges
Level
Expert
Scored
Checked automatically
Answers
Agent task
Tasks
Kept private
Strongest model tested
Claude Opus 5.5 #4 on the capability index
Last model added
October 2026
Scores refreshed
November 12th, 2026
8
Entries on its own leaderboard, 6 models we track

Claude Opus 5.5 leads with 87.7%

Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.

  1. 87.7 Claude Opus 5.5
  2. 80.9 gpt-6-sol-daybreak
  3. 72.7 DeepSeek V4.1 Flash 1017
  4. 63.2 Kimi K3.1 Pro
  5. 61.8 meta-wonder
  6. 54.4 Grok 4.7
  7. 37.6 Kimi K3
  8. 20.0 GLM-5.3-Flash
  1. 1 Claude Opus 5.5 87.7%
  2. 2 gpt-6-sol-daybreak 80.9%
  3. 3 DeepSeek V4.1 Flash 1017 72.7%
  4. 4 Kimi K3.1 Pro 63.2%
  5. 5 meta-wonder 61.8%
  6. 6 Grok 4.7 54.4%
  7. 7 Kimi K3 37.6%
  8. 8 GLM-5.3-Flash 20%

Source: Official CyberOffense leaderboard

#EntryScoreSettingDateSource
1 Claude Opus 5.5 87.7% Pi Sep 2026 Official leaderboard
2 gpt-6-sol-daybreak 80.9% Pi Official leaderboard
3 DeepSeek V4.1 Flash 1017 72.7% Pi Oct 2026 Official leaderboard
4 Kimi K3.1 Pro 63.2% Pi Oct 2026 Official leaderboard
5 meta-wonder 61.8% Pi Official leaderboard
6 Grok 4.7 54.4% Pi Sep 2026 Official leaderboard
7 Kimi K3 37.6% Pi Jul 2026 Official leaderboard
8 GLM-5.3-Flash 20% Pi Aug 2026 Official leaderboard

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

Anthropic
Own leaderboard Passed 85% in Oct 2026, by Claude Opus 5.5
Claude Opus 5.5 87.7%

Source: Official CyberOffense leaderboard

Claude Opus 5.5 on the official leaderboard passed 85% in September 2026

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) Claude Opus 5.5 Grok 4.7 Kimi K3 0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) Claude Opus 5.5 Grok 4.7

Source: Official CyberOffense leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

More in Cybersecurity

All 115 in Cybersecurity
  • ExploitGym In use 869 real-vulnerability exploit-development tasks (userspace, V8, Linux kernel). Success means capturing a flag through the intended vulnerability. GPT-6 Astra 42.4%
  • ExploitBench In use 41 recent V8 JavaScript-engine N-day vulnerabilities. Progress measured by 16 grader-verified capability flags across five tiers. GPT-6 Astra 100%
  • CyberBench v1.1 (Vals) In use Agents work on OSS-Fuzz memory-safety bugs from ARVO images in two tracks. They write a proof-of-concept input that crashes the target, or patch the source so the crash stops and tests still pass. Muse Wonder 80.4%
  • SRE-Bench In use Software reverse engineering without source: 262 binary instances from 19 private programs, each with six graded objectives. GPT-6 Astra 56.9%
  • CyScenarioBench In use Scenario-based long-horizon security challenges in sandboxes, organized in task, scenario and campaign tiers. Evaluation set kept private. Gemini 4 Argon 100%
  • ExploitBench - Internal Port In use Agents develop working exploits for recently disclosed vulnerabilities to limit contamination. Scored as arbitrary code execution success rate. GPT-6 Cyber 42%