Benchmark Almanac

Sandbox Bench

GPT-6 Astra 45.5% leads since Sep 2026 Only in system cards

Agents find and exploit vulnerabilities in runtimes, file format parsers and egress proxies across 22 challenges. Success means recovering the protected flag.

In-house OpenAI evaluation. Most targets are undisclosed zero days.

Made by OpenAI Released Sep 2026 Finding and exploiting real bugs Tasks kept private Last in a system card Oct 2026 Last model added Oct 2026
Level
Research frontier Researchers, on open problems or tasks that take days
Scored
Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
Answers
Agent task Many steps in a terminal, browser, desktop, app, game or robot.
4
Models scored in lab cards
4
System cards report it, from 1 lab

GPT-6 Astra leads with 45.5%

Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.

  1. 45.5 GPT-6 Astra
  2. 22.7 GPT-6.1 Terra
  3. 4.50 GPT-6 Luna
  4. 4.50 GPT-6 Sol
  1. 1 GPT-6 Astra 45.5%
  2. 2 GPT-6.1 Terra 22.7%
  3. 3 GPT-6 Luna 4.50%
  4. 4 GPT-6 Sol 4.50%
#ModelScoreSettingDateSource
1 GPT-6 Astra 45.5% Sep 2026 Model card
2 GPT-6.1 Terra 22.7% Oct 2026 Model card
3 GPT-6 Luna 4.50% Sep 2026 Model card
4 GPT-6 Sol 4.50% Sep 2026 Model card

Number one over time

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

OpenAI
Lab claims 1 leader
GPT-6 Astra 45.5%

Scores over time

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

Reported by the lab in its model card
0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) GPT-6 Astra 0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) GPT-6 Astra

Labs report it only in system cards

We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark. Some system cards do, below.

4 system cards report it, from 1 lab

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
26 Oct 2026 GPT-6.1 Terra system card addendum 22.7 Sandbox Bench, 5 of 22 targets succeeded (Figure 28, 9.0.2.3) Cyber risk
22 Sep 2026 GPT-6 Sol system card 4.5 Sandbox Bench, 1 of 22 targets (A.8.1.2.3) Cyber risk
22 Sep 2026 GPT-6 Luna system card 4.5 Sandbox Bench, 1 of 22 targets (A.8.1.2.3) Cyber risk
3 Sep 2026 GPT-6 Astra system card 45.5 Sandbox Bench, 10 of 22 targets (10.1.2.4) Cyber risk

More in Cybersecurity

All 111 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%
  • CWE-bench In use Agents audit a repository and write a patch for a security flaw, without being told where the bug is. Covers six programming languages. Gemini 4 Carbon 71%
  • Real-world vulnerability discovery In use Agents search popular open-source codebases to rediscover about 1,200 recent confirmed vulnerabilities across many languages. Reports share discovered. Gemini 4 Argon 86%
  • Artificial Analysis Cyber Index In use Agents do enterprise cyber defense tasks from three benchmarks, CWE-Bench-AA, DeepsecBench-AA and CyberGym-E2E-AA, combined into one success rate. Mistral Large 4 (Le Chonk, open weights) 51%
  • Big Sleep Evaluation (Google) In use Agents search large real codebases such as Chrome and Safari for critical, hard to find vulnerabilities. Scored by pass@1 success rate, run without safety guardrails. Gemini 3.5 Flash Cyber 72%