Hacker Blogs
Work in progress: 181,232 of 181,233 sites checked so far, more every week · updated 2 Nov 2026Missing a blog? Tell us

Top all-time · #9944

R

RUNTA Blog

Company blog · runta.com/blogaiprogramming

RUNTA Blog is the company blog at runta.com, writing about ai and programming. Since 2026, 8 different Hacker News readers have linked it 9 times, most often “Dive into FrontierHarness Eval: Claude Code cost 5.6× for the same pass rate”. It is still writing: the last post was on 2026-10-01, and you can follow it by feed.

Visit the blog
9mentions?
8HN people?
2026–2026first – last linked?

Too few sampled comments (3) to say how often HN recommends it.?

Why HN links it

HN commenters link Runta's FrontierHarness Eval posts, which compare nine agent harnesses on the same model by cost per pass, with code and tasks on GitHub. Comments are questions and links, not verdicts.

Summarised by Claude from a sample of 3 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.

Runta cuts what agents waste, bounds what they can reach, and records everything they did.

Most-linked posts?

  1. Dive into FrontierHarness Eval: Claude Code cost 5.6× for the same pass rate
    5 people · 5 mentions · first linked 2026-09-02
  2. Agents Aren't Software
    2 people · 2 mentions · first linked 2026-08-06

What Hacker News says?

  1. Edward40 · 2026-09-02
    Good points! That’s a known limitation of our v1.0 benchmark with Claude Code. For v1.1, we're expanding both harnesses and models to evaluate the full harness × model matrix. This will highlight interaction effects (showing which harnesses suit which models best) and let us publish a full compatibility matrix. We shared more details on our roadmap in our blog post ( https://runta.com/blog/introducing-frontierharnes…

    Comment linking /blog/introducing-frontierharness-eval/

    on “Show HN: FrontierHarness Eval – 9 harness, same model, cost per pass varies 17x”
  2. ADD-SP · 2026-09-09

    Submitted Dive into FrontierHarness Eval: Claude Code cost 5.6× for the same pass rate as a story · 6 points

    HN discussion
  3. molticrystal · 2026-09-02
    Their code and evalations are linked to on the blog post accompanying the results: Blog Post: https://runta.com/blog/introducing-frontierharness-eval/ The tasks and methodology: https://github.com/runta-dev/frontier-harness-eval

    Comment linking /blog/introducing-frontierharness-eval/

    on “Show HN: FrontierHarness Eval – 9 harness, same model, cost per pass varies 17x”
  4. shiqimei · 2026-08-13

    Submitted Runta SSH access for agent runtimes as a story · 6 points

    HN discussion
  5. xieydd · 2026-09-09
    Given the [K3 setup]( https://runta.com/blog/introducing-frontierharness-eval/ ), how much of the cost gap comes from Claude Code’s harness design versus its fit with the model and caching backend?

    Comment linking /blog/introducing-frontierharness-eval/

    on “Dive into FrontierHarness Eval: Claude Code cost 5.6× for the same pass rate”
  6. ADD-SP · 2026-09-22

    Submitted Errand – open-source Grok Bot and Muse alternative, built in a week as a story · 4 points

    HN discussion
  7. redman25 · 2026-09-03

    Submitted FrontierHarness Eval benchmark. Pi is on the Pareto as a story · 3 points

    HN discussion
  8. jinqueeny · 2026-08-16

    Submitted Agents Aren't Software as a story · 3 points

    HN discussion
  9. guanlan · 2026-08-06

    Submitted When a long-running agent crashes, what does recovery mean? as a story · 1 points

    HN discussion

Search all of HN for runta.com