Vals AI is an independent AI model evaluation firm led by co-founder and CEO Rayan Krishnan, assessing models from companies such as OpenAI and Anthropic. Recent coverage includes its finding that DeepSeek V4.1 Flash outperformed GPT 6 Sol, tested via the Daybreak trusted access program, on a multi-step offensive cybersecurity task, and its research showing AI agent teams cost up to 5.1x more than solo agents for little quality gain.
Vals AI: DeepSeek V4.1 Flash beats GPT 6 Sol on cybersecurity taskVals AI found that DeepSeek V4.1 Flash outperformed GPT 6 Sol on a multi-step offensive cybersecurity task. Sol was tested via the Daybreak trusted access program.
Research: AI agent teams cost up to 5.1x more for tiny quality gainsVals AI research finds AI agent teams barely outperform solo agents while costing up to 5.1x more; only one of four tests with GPT-6 Sol and Claude Opus 5.5 showed measurable gains. Anthropic data also shows quality plateauing beyond ten agents as token costs keep rising.
Can independent testing make AI safer?Rayan Krishnan, co‑founder and CEO of Vals AI, says investment in independent testing lags model capability gains. Vals evaluates models from firms like OpenAI and Anthropic, sees early signs of recursive self‑improvement, but notes models remain well behind top human researchers.
The A.I. Daily