Benchmark Almanac

ExploitBench - Internal Port

GPT-6 Cyber 42% leads since November 2026
In use

Agents develop working exploits for recently disclosed vulnerabilities to limit contamination. Scored as arbitrary code execution success rate.

OpenAI internal port of ExploitBench with June to August 2026 vulnerabilities.

Made by
OpenAI
Released
September 2026
Kind
Finding and exploiting real bugs
Level
Research frontier
Scored
Checked automatically
Answers
Agent task
Tasks
Kept private
Strongest model tested
GPT-6.1 Sol #5 on the capability index
Last in a launch post
November 2026
Last model added
November 2026
5
Models scored in lab cards
1
Launch post lists it, from 1 lab
6
System cards report it, from 1 lab

GPT-6 Cyber leads with 42%

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. 42.0 GPT-6 Cyber
  2. 24.5 GPT-6.1 Terra
  3. 21.5 GPT-6.1 Sol
  4. 5.50 GPT-6 Sol
  5. 0.0 GPT-6 Luna
  1. 1 GPT-6 Cyber 42%
  2. 2 GPT-6.1 Terra 24.5%
  3. 3 GPT-6.1 Sol 21.5%
  4. 4 GPT-6 Sol 5.50%
  5. 5 GPT-6 Luna 0%

Source: Labs’ model cards

#ModelScoreSettingDateSource
1 GPT-6 Cyber 42% Nov 2026 Lab model card
2 GPT-6.1 Terra 24.5% Oct 2026 Lab model card
3 GPT-6.1 Sol 21.5% Sep 2026 Lab model card
4 GPT-6 Sol 5.50% Sep 2026 Lab model card
5 GPT-6 Luna 0% Sep 2026 Lab model card

4 models have held number one in labs’ own numbers since its release in September 2026

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

OpenAI
Lab claims 4 leaders
GPT-6 Cyber 42%

Source: Labs’ model cards

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 Best so far
0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) GPT-6 Cyber GPT-6.1 Terra GPT-6 Sol 0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) GPT-6 Cyber GPT-6.1 Terra

Source: Labs’ model cards

One lab put it in a launch post in the last six months

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.

OpenAI
DateLaunch postScoreAs the post writes it
Nov 5, 2026 GPT-6 Cyber 42 ExploitBench Internal Port, mean arbitrary code-execution success rate across V8 and Chrome targets, June to August 2026

6 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
Nov 5, 2026 GPT-6 Cyber system card 42 ExploitBench Internal Port, mean arbitrary code-execution success rate across V8 and Chrome targets, June to August 2026 Cyber risk
Oct 26, 2026 GPT-6.1 Terra system card addendum 24.5 ExploitBench - Internal Port (June–August 2026), arbitrary code-execution success rate (9.0.2.2) Cyber risk
Sep 29, 2026 GPT-6.1 Sol system card addendum 21.5 ExploitBench - Internal Port (June–August 2026), arbitrary code-execution success rate (9.1.2.2) Cyber risk
Sep 22, 2026 GPT-6 Sol system card 5.5 ExploitBench - Internal Port (June–August 2026), maximum success rate (A.8.1.2.2) Cyber risk
Sep 22, 2026 GPT-6 Luna system card 0 ExploitBench - Internal Port (June–August 2026), text says no challenge completed (A.8.1.2.2) Cyber risk
Sep 3, 2026 GPT-6 Astra system card ExploitBench - Internal Port (June–August 2026), success rate vs output tokens, chart without printed numbers (Figure 47) Cyber risk

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%
  • MiMo Cyber Bench In use Agents do in-house cybersecurity tasks that add to public vulnerability reproduction and exploitation benchmarks. The report gives no task count or grading details. MiMo-V2.6 Pro 81.7%