CWE-bench v1
PulseAugur coverage of CWE-bench v1 — every cluster mentioning CWE-bench v1 across labs, papers, and developer communities, ranked by signal.
-
Google releases Gemini 4 Argon with 1M token limit for cyber defenders
Google has announced Gemini 4 Argon, a new large language model designed for long-horizon tasks, particularly in software engineering, enterprise knowledge work, and cybersecurity. The model is being released in a phase…
-
Google releases guarded Gemini 4 Argon; OpenAI thwarts data extraction attempt
Google has launched its new frontier model, Gemini 4 Argon, initially restricting access to trusted cybersecurity professionals through its Fairwind Program. This model is designed for advanced cybersecurity capabilitie…
-
Gemini 4 Argon leads benchmarks but remains inaccessible to most
Google's Gemini 4 Argon model has demonstrated superior performance across various benchmarks, outperforming OpenAI's GPT-6 Astra and Anthropic's Claude Opus 5.5 and Claude Fable 5.1 in many categories, particularly in …
-
Google DeepMind launches Gemini 4 Argon with 1M output tokens · 10 sources tracked
Google DeepMind has unveiled Gemini 4 Argon, a new frontier model designed for complex, long-horizon tasks in coding, enterprise knowledge work, and cybersecurity. A key advancement is its ability to generate up to 1 mi…
-
Google launches Gemini 4 Argon, prioritizing cybersecurity partners
Google has launched Gemini 4 Argon, its latest advanced AI model, focusing on cybersecurity and complex coding tasks. The model is initially available to select cybersecurity partners through Google's Fairwind Program, …