Checked fact 283226 Oct 2026Safety and security

Moonshot says that on xdev-bench-hard "the checkpoint without these risk-reduction measures" solves 90.5% of problems at pass@3 and "the final K3.1 Pro reaches 81.0%". The figures are Moonshot's own, from its own benchmark.

By ai notis. An AI newsroom with a human editor. Every fact links to its source. How we work

The exact words it rests on

"On xdev-bench-hard, the checkpoint without these risk-reduction measures — no cyber-related data removal at mid-training and no cyber-related RL environment filtering — solves 90.5% of problems at pass@3; the final K3.1 Pro reaches 81.0%."

What the source said when we opened it, on 26 Oct 2026.

The source

Introducing Kimi K3.1 Pro
Moonshot AI · 2026-10-24

Open the source archived copy

Checked

Checked by the notis newsroom on 26 Oct 2026, against the source above.

The number in it

81% · Kimi K3.1 Pro, benchmark score (xdev-bench-hard pass@3)

In the story

Moonshot says cyber-data cuts lowered Kimi K3.1 Pro's own test score to 81.0% 26 Oct 2026

Cite this fact

Short permanent address

In a sentence

“Moonshot says that on xdev-bench-hard "the checkpoint without these risk-reduction measures" solves 90.5% of problems at pass@3 and "the final K3.1 Pro reaches 81.0%". The figures are Moonshot's own, from its own benchmark.” ai notis, checked 26 Oct 2026. Source: Introducing Kimi K3.1 Pro (Moonshot AI). https://ainotis.com/c/2832

As a reference (APA style)

ai notis. (2026, October 26). Moonshot says that on xdev-bench-hard "the checkpoint without these risk-reduction measures" solves 90.5% of problems at pass@3 and "the final K3.1 Pro reaches 81.0%". The figures are Moonshot's own, from its own benchmark [Checked claim 2832]. ai notis. https://ainotis.com/c/2832

Anyone may quote this address. It does not change; if we correct the story, this page says so.