NUWA 
DeepSeek
DeepSeek-V4-Flash
Published evidence on capabilities, behavioral tendencies and observed actions.
- First public availability
- 2026-04-24
- Parameters
- 284B · 13B active · MoE
- Metadata updated
- 2026-10-27
Evidence overview
Developer documentation is linked below. Structured risk results have not yet been extracted for this exact model record.
NUWA curates and interprets these sources. Experiments are attributed to their original executors; this page does not claim NUWA has independently replicated them.
MODEL MONITORING
Representative benchmark coverage
Results by protocol and configuration; gaps show records awaiting collection or linkage.
AgentCyberRange →
2 measurements
Cybench →
No eligible measurement linked to this model version.
CyberGym →
11 measurements
All conditions →METR Time Horizons →
No eligible measurement linked to this model version.
SEC-bench Pro →
No eligible measurement linked to this model version.
MASK →
No eligible measurement linked to this model version.
Attempt to Persuade Eval (APE) →
No eligible measurement linked to this model version.
DeceptionBench →
No eligible measurement linked to this model version.
PersuSafety →
No eligible measurement linked to this model version.
All linked measurements
Each result retains its protocol and unit.
13 observations
| System / model | Result | Condition status | Task / conditions | Evaluator / report | Dates |
|---|---|---|---|---|---|
| DeepSeek-V4-Flash · DeepSeek AgentDeepSeek-V4-Flash | 76.7%CyberGym · Success rate | Incomplete | level1score_10DeepSeek Agent | Not disclosedCyberGym official leaderboard and machine-data extraction | Release 2026-04-24Report — |
| DeepSeek-V4-Flash · NASHDeepSeek-V4-Flash | 91.4%CyberGym · Success rate | Incomplete | level1score_10NASH | Not disclosedCyberGym official leaderboard and machine-data extraction | Release 2026-04-24Report — |
| DeepSeek-V4-Flash · PwnBotDeepSeek-V4-Flash | 93.1%CyberGym · Success rate | Incomplete | level1score_10PwnBot | Not disclosedCyberGym official leaderboard and machine-data extraction | Release 2026-04-24Report — |
| DeepSeek-V4-Flash · Ostorlab NeutronDeepSeek-V4-Flash | 97.1%CyberGym · Success rate | Incomplete | level1score_10Ostorlab Neutron | Not disclosedCyberGym official leaderboard and machine-data extraction | Release 2026-04-24Report — |
| DeepSeek-V4-Flash · TriLaneDeepSeek-V4-Flash | 88.9%CyberGym · Success rate | Incomplete | level1score_10TriLane | Not disclosedCyberGym official leaderboard and machine-data extraction | Release 2026-04-24Report — |
| DeepSeek-V4-Flash · VARAS-OneMind \(智安·宇盾\)DeepSeek-V4-Flash | 97%CyberGym · Success rate | Incomplete | level1score_10VARAS-OneMind \(智安·宇盾\) | Not disclosedCyberGym official leaderboard and machine-data extraction | Release 2026-04-24Report — |
| DeepSeek-V4-Flash · GusionDeepSeek-V4-Flash | 87.1%CyberGym · Success rate | Incomplete | level1score_10Gusion | Not disclosedCyberGym official leaderboard and machine-data extraction | Release 2026-04-24Report — |
| DeepSeek-V4-Flash · FangcunCyberDeepSeek-V4-Flash | 91.3%CyberGym · Success rate | Incomplete | level1score_10FangcunCyber | Not disclosedCyberGym official leaderboard and machine-data extraction | Release 2026-04-24Report — |
| DeepSeek-V4-Flash · RO0T AgentDeepSeek-V4-Flash | 94.2%CyberGym · Success rate | Incomplete | level1score_10RO0T Agent | Not disclosedCyberGym official leaderboard and machine-data extraction | Release 2026-04-24Report — |
| DeepSeek-V4-Flash · Sangfor AIDeepSeek-V4-Flash | 93.2%CyberGym · Success rate | Incomplete | level1score_10Sangfor AI | Not disclosedCyberGym official leaderboard and machine-data extraction | Release 2026-04-24Report — |
| DeepSeek-V4-Flash · Whitzard \(白泽\)DeepSeek-V4-Flash | 91.2%CyberGym · Success rate | Incomplete | level1score_10Whitzard \(白泽\) | Not disclosedCyberGym official leaderboard and machine-data extraction | Release 2026-04-24Report — |
| DeepSeek-V4-Flash / Claude Code 2.1.226DeepSeek-V4-Flash | 14.19%AgentCyberRange · Success rate (Pass@3) | Incomplete | Web exploitationlevel0Claude Code | Not disclosedAgentCyberRange: Evaluate Agent Cyberattack Capabilities in Enterprise-like Cyber Ranges | Release 2026-04-24Report — |
| DeepSeek-V4-Flash / Claude Code 2.1.226DeepSeek-V4-Flash | 28.46%AgentCyberRange · Success rate (Pass@3) | Incomplete | Post exploitationlevel0Claude Code | Not disclosedAgentCyberRange: Evaluate Agent Cyberattack Capabilities in Enterprise-like Cyber Ranges | Release 2026-04-24Report — |
Developer disclosures
DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence →
Developer documentation covering DeepSeek-V4-Pro, DeepSeek-V4-Flash.
Publication date not verifiedExternal assessments
CyberGym official snapshot →
Published evaluation results collected from CyberGym, retaining the reported systems and configurations.
Publication date not establishedAgentCyberRange official snapshot →
Published evaluation results collected from AgentCyberRange, retaining the reported systems and configurations.
Publication date not establishedEvidence by risk topic
Autonomy No structured result linked
A gap in this collection is not evidence of low risk. Consult the linked reports for unextracted findings and protocol scope.
Deception No structured result linked
A gap in this collection is not evidence of low risk. Consult the linked reports for unextracted findings and protocol scope.
Cybersecurity 13 protocol results
CyberGym · level1 · score_10 →
External evaluation · Executed by
- Success rate
- 91.4%
Per-result sample size not disclosed
CyberGym · level1 · score_10 →
External evaluation · Executed by
- Success rate
- 93.1%
Per-result sample size not disclosed
CyberGym · level1 · score_10 →
External evaluation · Executed by
- Success rate
- 97.1%
Per-result sample size not disclosed
CyberGym · level1 · score_10 →
External evaluation · Executed by
- Success rate
- 88.9%
Per-result sample size not disclosed
CyberGym · level1 · score_10 →
External evaluation · Executed by
- Success rate
- 97%
Per-result sample size not disclosed
CyberGym · level1 · score_10 →
External evaluation · Executed by
- Success rate
- 87.1%
Per-result sample size not disclosed
CyberGym · level1 · score_10 →
External evaluation · Executed by
- Success rate
- 91.3%
Per-result sample size not disclosed
CyberGym · level1 · score_10 →
External evaluation · Executed by
- Success rate
- 94.2%
Per-result sample size not disclosed
CyberGym · level1 · score_10 →
External evaluation · Executed by
- Success rate
- 76.7%
Per-result sample size not disclosed
CyberGym · level1 · score_10 →
External evaluation · Executed by
- Success rate
- 93.2%
Per-result sample size not disclosed
CyberGym · level1 · score_10 →
External evaluation · Executed by
- Success rate
- 91.2%
Per-result sample size not disclosed
AgentCyberRange · Web exploitation · pass3 →
External evaluation · Executed by
- Success rate (Pass@3)
- 14.19%
Per-result sample size not disclosed
AgentCyberRange · Post exploitation · pass3 →
External evaluation · Executed by
- Success rate (Pass@3)
- 28.46%
Per-result sample size not disclosed
CBRN No structured result linked
A gap in this collection is not evidence of low risk. Consult the linked reports for unextracted findings and protocol scope.
Disinformation No structured result linked
A gap in this collection is not evidence of low risk. Consult the linked reports for unextracted findings and protocol scope.
Autonomous physical harm No structured result linked
A gap in this collection is not evidence of low risk. Consult the linked reports for unextracted findings and protocol scope.
Source review and evidence gaps
CyberGym →
Published evaluation results collected from CyberGym, retaining the reported systems and configurations.
The site reports submitted systems under different configurations. Per-row task counts, budgets and execution dates are not fully disclosed. Undefined metrics and unresolved identity rows are excluded. Raw JSON fractions are shown as percentages using ×100 and one-decimal rounding, matching the official renderer.
AgentCyberRange →
Published evaluation results collected from AgentCyberRange, retaining the reported systems and configurations.
Web exploitation and post-exploitation are separate evaluations. Pass@3 retains the site’s definition; row-specific budgets, sample counts and execution dates are not disclosed.
DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence →
Source-specific extraction of DeepSeek V4 Technical Report. Findings and measurements retain their original attribution.
Source transcription is not experimental replication. Unknown dates, evaluator roles and configuration details remain unknown. Source groups are not verified rankings.
Missing scores may reflect unextracted sections, undisclosed configurations, unresolved identity or an absent experiment. These are separate research questions; none implies a zero result.
Interpretation and open questions
Differences between reports can reflect task selection, model endpoints, prompts, budgets, judges and publication dates. Evidence is grouped by the protocol actually used; no model-wide safety grade is inferred.
References
- [1] deepseek-v4-flash release ; News article title, dated heading and availability
- [2] deepseek-v4-flash developer model card ; Introduction; model comparison table
- [3] DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
- [4] CyberGym official leaderboard and machine-data extraction ; /level1/9/score_10; /level1/11/score_10; /level1/13/score_10
- [5] AgentCyberRange: Evaluate Agent Cyberattack Capabilities in Enterprise-like Cyber Ranges ; routes-B-0Mubrq.js::QR/6/level0/pass3; routes-B-0Mubrq.js::$R/6/level0/pass3