cd /news/artificial-intelligence/claude-did-best-on-a-new-benchmark-f… · home › topics › artificial-intelligence › article
[ARTICLE · art-125246] src=thenewstack.io ↗ pub= topic=artificial-intelligence verified=true sentiment=· neutral

Claude did best on a new benchmark for ‘agents that build agents’. It still passed fewer than a quarter of the tests.

Anthropic's Claude topped a new benchmark for AI agents that build other agents, yet it still passed fewer than a quarter of the tests, according to The New Stack. The benchmark evaluates models' ability to create functional agents, highlighting the current limitations of even the best-performing AI systems.

by read1 min views1 publishedSep 9, 2026

AI models now power all manner of agents, from coding assistants that write and debug software to customer service systems

The post Claude did best on a new benchmark for ‘agents that build agents’. It still passed fewer than a quarter of the tests. appeared first on The New Stack.

source & further reading
thenewstack.io — original article How much control should AI get? A CISO roundtable takes on SOC autonomy After nine years as HashiCorp CEO, Dave McJannet now wants to “unblock” enterprise AI agents Claude Fable 5.1 vs. Fable 5: On real work, I couldn’t tell them apart.
── more in #artificial-intelligence 4 stories · sorted by recency
── more on @anthropic 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/claude-did-best-on-a…] indexed:0 read:1min 2026-09-09 · —