ExploitGym
Created by Berkeley RDI, 11 May 2026 · Higher is better · Website
Settings
| Task set | Scoring | Mitigations | Reported by |
|---|---|---|---|
| ExploitGym 1.0 (869 tasks) | Score | Mitigations off | |
| ExploitGym 1.0 (869 tasks) | Score | Mitigations on | |
| ExploitGym 1.0 (869 tasks) | Score | Not specified | |
| ExploitGym 1.0 (869 tasks) | Successful exploits | Mitigations off | |
| ExploitGym 1.0 (869 tasks) | Successful exploits | Not specified |
Bold: Benchmark default based on Paper and Leaderboard · Download image
Results
Show
- OpenAI
- Other labs
- Reported in a later post or card
- OpenAI
- Other labs
- Reported in a later post or card
| Model | Harness and effort | Score | Source |
|---|---|---|---|
| GPT-6-Cyber5 Nov 2026 | Harness not specifiedEffort not specified | 48.5% | GPT‑6 Cyber: Advancing the frontier of aligned cybersecurity agentsChecked A person checked this number against its source on 20 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced. |
| GPT-6 Astra3 Sep 2026 | Harness not specifiedEffort not specified | 42.4% | GPT-6 Astra: A new generation of intelligenceCorroborated The same number is printed in 4 other documents. The first 3 are linked: 1 2 3no 6h time limit, output-token cap per the system card · "v1" implementation with runtime package installation disabled and the environment offline, per the system card · Daybreak Blue access, per the GPT-6 Cyber post · without production safeguardsOur capture |
| GPT-6 Astra3 Sep 2026 | Harness not specifiedEffort not specified | 0.0% | GPT‑6 Cyber: Advancing the frontier of aligned cybersecurity agentsChecked A person checked this number against its source on 20 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.strict supplied-vulnerability pass@1 · safeguards enabled · reported later · another reportOur capture |
| GPT-6.1 Sol29 Sep 2026 | Harness not specifiedEffort not specified | 35.1% | GPT-6.1 Sol System Card AddendumCorroborated The same number is printed in 1 other document: 1per attempt |
| GPT-5.6 Sol9 Jul 2026 | Harness not specifiedEffort not specified | 33.7% | GPT-5.6: Frontier intelligence that scales with your ambitionCorroborated The same number is printed in 1 other document: 16h time limit, timed on OpenAI's faster alpha API and rescaled to the public API · reduced safeguardsOur capture |
| GPT-5.6 Sol9 Jul 2026 | Harness not specifiedEffort not specified | 24.9% | GPT-5.6: Frontier intelligence that scales with your ambitionChecked A person checked this number against its source on 20 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.2h cap, timed on OpenAI's faster alpha API and rescaled to the public API · reduced safeguards · another report |
| GPT-5.6 Sol9 Jul 2026 | Harness not specifiedEffort not specified | 30.3% | GPT-6 Astra: A new generation of intelligenceCorroborated The same number is printed in 3 other documents: 1 2 3no 6h time limit · Daybreak Blue access, per the GPT-6 Cyber post · system-level guardrails removed, with safeguards tailored to authorized defensive security work, per the GPT-5.6-Cyber post · reported later · another reportOur capture |
| GPT-5.6 Sol9 Jul 2026 | Harness not specifiedhighest publicly available reasoning level | 30.8% | Expanding Daybreak as the Cyber Defense Window NarrowsChecked A person checked this number against its source on 20 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.Daybreak Blue access · system-level guardrails removed, with safeguards tailored to authorized defensive security work · OpenAI's internal implementation in security-hardened, isolated environments · reported later · another report |
| GPT-5.6 Sol9 Jul 2026 | Harness not specifiedhighest publicly available reasoning level | 0.0% | Expanding Daybreak as the Cyber Defense Window NarrowsChecked A person checked this number against its source on 20 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.safeguards enabled · OpenAI's internal implementation in security-hardened, isolated environments · reported later · another report |
| GPT-5.6-Cyber10 Aug 2026 | Harness not specifiedhighest publicly available reasoning level | 32.1% | Expanding Daybreak as the Cyber Defense Window NarrowsCorroborated The same number is printed in 1 other document: 1Daybreak Red access · OpenAI's internal implementation in security-hardened, isolated environments |
| GPT-5.6 Terra9 Jul 2026 | Harness not specifiedEffort not specified | 23.2% | GPT-5.6: Frontier intelligence that scales with your ambitionChecked A person checked this number against its source on 20 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.timed on OpenAI's faster alpha API and rescaled to the public API · reduced safeguardsOur capture |
| GPT-6 Sol22 Sep 2026 | Harness not specifiedEffort not specified | 22.1% | GPT-6.1 Sol System Card AddendumCorroborated The same number is printed in 1 other document: 1reported later |
| GPT-5.5-Cyber7 May 2026 | Harness not specifiedhighest publicly available reasoning level | 21.2% | Expanding Daybreak as the Cyber Defense Window NarrowsChecked A person checked this number against its source on 20 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.Daybreak Red access · OpenAI's internal implementation in security-hardened, isolated environments · reported later |
| Kimi K3.1 Pro24 Oct 2026 | KimiCodemax effort | 20.7% | Kimi K3.1 Pro Tech Blog: Advancing Open IntelligenceChecked A person checked this number against its source on 20 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.6h wall-clock budget per taskOur capture |
| DeepSeek-V4.1-Flash10 Sep 2026 | Harness not specifiedEffort not specified | 15.3% | DeepSeek-V4.1-Flash: Smarter, Faster, More EfficientChecked A person checked this number against its source on 20 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.Source image |
| GPT-5.523 Apr 2026 | Harness not specifiedEffort not specified | 15.1% | GPT-5.6: Frontier intelligence that scales with your ambitionChecked A person checked this number against its source on 20 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.2h cap, timed on OpenAI's faster alpha API and rescaled to the public API · reduced safeguards · reported laterOur capture |
| GPT-5.6 Luna9 Jul 2026 | Harness not specifiedEffort not specified | 12.4% | GPT-5.6: Frontier intelligence that scales with your ambitionChecked A person checked this number against its source on 20 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.timed on OpenAI's faster alpha API and rescaled to the public API · reduced safeguardsOur capture |
| GPT-6 Luna22 Sep 2026 | Harness not specifiedEffort not specified | 11.6% | GPT-6.1 Sol System Card AddendumCorroborated The same number is printed in 1 other document: 1reported later |
| DeepSeek-V4-Pro-081313 Aug 2026 | Harness not specifiedEffort not specified | 5.4% | DeepSeek-V4.1-Flash: Smarter, Faster, More EfficientChecked A person checked this number against its source on 20 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.reported laterSource image |
| Muse Spark 1.19 Jul 2026 | Harness not specifiedhigh reasoning effort | 0.8% | Muse Spark 1.1 Evaluation ReportChecked A person checked this number against its source on 20 Nov 2026. This confirms the number matches what the lab published. It doesn't confirm that the score was reproduced.pass@1 · Muse Spark 1.1 Helpful, no system prompt |