Wire flash
DeepSeek V4.1 Flash 1017 tops openly-available cyber index at 57%
Editorial responsibility
- No named human review is recorded for this page.
- Source reporting is collected, normalized, translated or condensed automatically when needed.
- Automatically published source-backed update
DeepSeek V4.1 Flash 1017, an update to DeepSeek's flagship open-weights model, is now the top openly-available model on the Artificial Analysis Cyber Index with a score of 57%, according to Artificial Analysis. The index measures agentic cyber defense tasks across CWE-Bench-AA, DeepsecBench-AA, and CyberGym-E2E-AA, evaluated mainly through public APIs. DeepSeek V4.1 Flash 1017 ranks #2 overall, 12 points behind the trusted-access model GPT-6 Sol (Daybreak Blue, max) at 69%, and one point ahead of Grok 4.7 (xhigh) and MiMo-V2.6-Pro. It improves 16 points over its predecessor, driven by a 53-point gain on CyberGym-E2E-AA from 23% to 76%. At $0.42 per task, it sits on the capability-versus-cost Pareto frontier, over 2x MiMo-V2.6-Pro's $0.18 but 4x cheaper than GPT-6 Sol's $1.77.
Source report
DeepSeek V4.1 Flash 1017 is the new #1 openly-available model in the Artificial Analysis Cyber Index. Other proprietary models may offer better cyber defense performance but are restricted access.
Overview
DeepSeek recently released DeepSeek V4.1 Flash 1017, an update to its flagship open-weights model, DeepSeek V4.1 Flash. With an Artificial Analysis Cyber Index score of 57%, DeepSeek V4.1 Flash 1017 is the #1 openly-available model for cyber defense capabilities.
About the Cyber Index
The Artificial Analysis Cyber Index measures capability on agentic cyber defense tasks and is composed of three benchmarks: CWE-Bench-AA, DeepsecBench-AA, and CyberGym-E2E-AA. The Cyber Index primarily evaluates models through their publicly accessible APIs, reflecting the cyber capabilities available to developers.
Many models, particularly proprietary models from OpenAI, Anthropic, and Google, may have stronger defensive cyber capabilities but encounter safety refusals that reduce their overall scores. GPT-6 Sol (Daybreak Blue, max), available through OpenAI's trusted-access program, leads the Index, scoring 32 points above its public counterpart due to fewer refusals. DeepSeek V4.1 Flash 1017 encounters no safety refusals.
Key Results
- Highest score among publicly-available models, #2 overall. DeepSeek V4.1 Flash 1017 (max) scores 57%, ahead of the previously leading publicly-available models Grok 4.7 (xhigh) and MiMo-V2.6-Pro by 1 point. It places #2 overall, with the trusted-access model GPT-6 Sol (Daybreak Blue, max) leading by 12 points with a score of 69%.
- 16-point improvement over its predecessor. DeepSeek V4.1 Flash 1017 improves on DeepSeek V4.1 Flash by 16 points, from 41% to 57%. This gain is driven by a substantial 53-point improvement on CyberGym-E2E-AA, from 23% to 76%. On CWE-Bench-AA and DeepsecBench-AA, it scores 69% and 25%, compared to 68% and 31% for DeepSeek V4.1 Flash (max).
- On the Capability vs. Cost Pareto frontier. With a Cost per Task of $0.42, DeepSeek V4.1 Flash 1017 costs more than 2x MiMo-V2.6-Pro (56%) at $0.18 per task, but is 4x cheaper than GPT-6 Sol (Daybreak Blue, max, 69%) at $1.77 per task.
Source
ArtificialAnlysNeutral / independent
Part of this Story
DeepSeek V4.1 Flash 1017 Tops Open Cyber Index at 57%, Scores 45 on Intelligence Index