Hacker News new | ask | show | jobs
by WithinReason 1 hour ago
> [...] beating Claude Code (32%) at roughly $0.17 per vulnerability found

Claude Code is an agent harness, not an LLM.

Claude is a brand (or group of LLMs), not an LLM.
3 comments
raincole 1 hour ago
Yes, and the article author is fully aware of that. Thank you for pointing out this small mistake though.
link
mkagenius 18 minutes ago
It looks like the author is specifically avoiding model's name, because results are really weird.

  Opus 4.8/4.7 scored 28%

  Opus 4.6 score 37%
So the author thought as let's not get into that just write Claude.
link
andriy_koval 11 minutes ago
many people think opus 4.6 was the best
link
tills13 33 minutes ago
It costs nothing to not be pedantic.
link
Onavo 1 hour ago
Claude code it's the only way to get access to the actual amortized cost of running a Claude-scale model. The consumer non-enterprise API is extremely expensive (with increasing marginal costs for the user and fat profit margins for Anthropic). If you want to approximate a State level attacker's cost where they can have the model on their own hardware, Claude Code is probably the best guess at the amortized cost.
link