Skip to content
LiveNext 20:10:23
topfive.ai
CoolingSafety

Vals benchmark: open DeepSeek model nears GPT-6 Sol Daybreak on cyber tasks at a fraction of the cost

Vals AI's CyberOffense benchmark, last updated Oct. 23, has DeepSeek V4.1 Flash 1017 averaging 16.3 of 21 objectives on Nova versus 15.7 for GPT-6 Sol Daybreak, though Daybreak led on Lighthouse, 27 of 31 to 21. Vals lists under $3 a run for DeepSeek and about $22 for Daybreak. The Decoder reported the comparison Nov. 7.

HEAT
OUTLETS
1
FIRST SEEN
LAST SIGNAL

Primary source

Vals AIvals.ai/benchmarks/cyber_offense

Coverage · 2 articles, oldest first

Saturday, November 7, 2026

  1. The Decoder

    The gap between open and closed AI models in cyber capabilities keeps shrinking

  2. Vals AIPrimary

    Vals benchmark: open DeepSeek model nears GPT-6 Sol Daybreak on cyber tasks at a fraction of the cost
Share
Share card: Vals benchmark: open DeepSeek model nears GPT-6 Sol Daybreak on cyber tasks at a fraction of the cost

Share this article

Copying the image…

Open any post or message and paste it in. The link to TopFive is printed on the card.

Post it on

X LinkedIn Hacker News Email
Download image
XLinkedInHacker NewsEmail
← The whole wire