Trendoday

Show HN: KradleBench – Multi-agent AI evals in Minecraft

1 comments

  • Points9
  • Comments7
–Rank now
#7Peak rank
2Days on list
249Score · 7d
Nov 18, 2026First seen
Rank · 7 days

On Hacker News

Hi HN - This is James and Kemal. In past lives we led Firebase and Tensorflow/JAX respectively. HN has been a particularly big part of my (James’) career and the trajectory I’ve taken would be meaningfully different without it. Thank you! Our team shipped a benchmark: KradleBench. We test AI in Minecraft Worlds. 1. The worlds are multi-agent. Humans and other AIs are in there too, often pursuing their own goals. Collaboration and conflict skills fall out of this property. 2. The world are…

Points
9
Comments
7
Submitted by
jamest
Posted
Nov 18, 2026
Site
kradle.ai

Story details from Hacker News ↗ · linked article on kradle.ai.

Trending history

Show HN: KradleBench – Multi-agent AI evals in Minecraft was first seen on Hacker News · Show HN on Nov 18, 2026. Since then it has appeared in 47 checks on 2 different days. Its best position was #7 of 30, reached on Nov 18, 20:52 UTC. It dropped off the list after Nov 19, 18:53 UTC.

Rank history

Position on Hacker News · Show HN at every check (lower is better).

#1#11#20#30 19:5201:3807:2313:0818:53 Nov 18, 19:52 UTC · #20Nov 18, 20:22 UTC · #10Nov 18, 20:52 UTC · #7Nov 18, 21:22 UTC · #7Nov 18, 21:52 UTC · #8Nov 18, 22:22 UTC · #8Nov 18, 22:52 UTC · #10Nov 18, 23:22 UTC · #10Nov 18, 23:52 UTC · #10Nov 19, 00:22 UTC · #10Nov 19, 00:52 UTC · #11Nov 19, 01:22 UTC · #11Nov 19, 01:52 UTC · #11Nov 19, 02:22 UTC · #11Nov 19, 02:52 UTC · #11Nov 19, 03:22 UTC · #11Nov 19, 03:52 UTC · #12Nov 19, 04:22 UTC · #11Nov 19, 04:52 UTC · #11Nov 19, 05:22 UTC · #11Nov 19, 05:52 UTC · #11Nov 19, 06:22 UTC · #12Nov 19, 06:53 UTC · #12Nov 19, 07:23 UTC · #12Nov 19, 07:53 UTC · #11Nov 19, 08:23 UTC · #11Nov 19, 08:53 UTC · #10Nov 19, 09:23 UTC · #10Nov 19, 09:53 UTC · #10Nov 19, 10:23 UTC · #11Nov 19, 10:53 UTC · #11Nov 19, 11:23 UTC · #10Nov 19, 11:53 UTC · #10Nov 19, 12:23 UTC · #10Nov 19, 12:53 UTC · #10Nov 19, 13:23 UTC · #11Nov 19, 13:53 UTC · #11Nov 19, 14:23 UTC · #13Nov 19, 14:53 UTC · #13Nov 19, 15:23 UTC · #13Nov 19, 15:53 UTC · #14Nov 19, 16:23 UTC · #12Nov 19, 16:53 UTC · #12Nov 19, 17:23 UTC · #12Nov 19, 17:53 UTC · #12Nov 19, 18:23 UTC · #12Nov 19, 18:53 UTC · #13

Last 14 days

Best rank per day (UTC).

Nov 7today

Metric history

Platform metric as reported by Hacker News.

0369 19:5201:3807:2313:0818:53 Nov 18, 19:52 UTC · 4Nov 18, 20:22 UTC · 6Nov 18, 20:52 UTC · 8Nov 18, 21:22 UTC · 8Nov 18, 21:52 UTC · 8Nov 18, 22:22 UTC · 8Nov 18, 22:52 UTC · 8Nov 18, 23:22 UTC · 8Nov 18, 23:52 UTC · 8Nov 19, 00:22 UTC · 8Nov 19, 00:52 UTC · 8Nov 19, 01:22 UTC · 8Nov 19, 01:52 UTC · 8Nov 19, 02:22 UTC · 8Nov 19, 02:52 UTC · 8Nov 19, 03:22 UTC · 8Nov 19, 03:52 UTC · 8Nov 19, 04:22 UTC · 8Nov 19, 04:52 UTC · 8Nov 19, 05:22 UTC · 8Nov 19, 05:52 UTC · 8Nov 19, 06:22 UTC · 8Nov 19, 06:53 UTC · 8Nov 19, 07:23 UTC · 8Nov 19, 07:53 UTC · 9Nov 19, 08:23 UTC · 9Nov 19, 08:53 UTC · 9Nov 19, 09:23 UTC · 9Nov 19, 09:53 UTC · 9Nov 19, 10:23 UTC · 9Nov 19, 10:53 UTC · 9Nov 19, 11:23 UTC · 9Nov 19, 11:53 UTC · 9Nov 19, 12:23 UTC · 9Nov 19, 12:53 UTC · 9Nov 19, 13:23 UTC · 9Nov 19, 13:53 UTC · 9Nov 19, 14:23 UTC · 9Nov 19, 14:53 UTC · 9Nov 19, 15:23 UTC · 9Nov 19, 15:53 UTC · 9Nov 19, 16:23 UTC · 9Nov 19, 16:53 UTC · 9Nov 19, 17:23 UTC · 9Nov 19, 17:53 UTC · 9Nov 19, 18:23 UTC · 9Nov 19, 18:53 UTC · 9

From the discussion

A few top-level comments from Hacker News, shortened. Read the full thread on Hacker News.

(vikrum from Kradle here, too) One of the most interesting things about this benchmark is that it encodes latency into the score as agents navigate the world where other agents are also doing things. Just because a model can score high when there is no time limit is not useful for many important workloads like active…

Florian from Kradle here - happy to answer any question! We're very excited about this release. We did not expect Astra 6.2 to score so high, especially with its current online reputation. Looks like it's better as an autonomous agent than a coding assistant!

Great bench! Question: from playing with the models why do you think developers seem to prefer Opus / Fable, whereas Astra does much better here? What do you think Astra is lacking for them?

Questions

Is Show HN: KradleBench – Multi-agent AI evals in Minecraft trending on Show HN right now?

Not at the moment. It was last on the list on Nov 19, 18:53 UTC, after 2 days of trending.

What is the highest rank Show HN: KradleBench – Multi-agent AI evals in Minecraft reached?

#7 on Hacker News · Show HN, on Nov 18, 20:52 UTC. Trendoday checks this list every 30 minutes.

How long has Show HN: KradleBench – Multi-agent AI evals in Minecraft been trending?

It was first seen on Nov 18, 2026 and has been on the list on 2 days (47 checks) since.

How much discussion did it get on Hacker News?

9 points and 7 comments; submitted by jamest on Nov 18, 2026.