The 12 October post says that in Anthropic's own test a pretrained Claude model trained on full reasoning transcripts from Claude Opus 4.8 scored 80.4% on SWE-bench Verified, against 71.8% when trained on summarised transcripts.

The exact words it rests on

We tested how well a pretrained Claude model performed on the SWE-bench Verified coding benchmark after it was trained on transcripts from Claude Opus 4.8. When training on the full transcripts, the distilled model achieved 80.4%; with only the summarized transcripts, it achieved 71.8%.

What the source said when we opened it, on 13 Oct 2026.

The source

Strengthening our defenses against distillation attacks
Anthropic · 2026-10-12

Open the sourcearchived copy

Checked

Checked by the notis newsroom on 13 Oct 2026, against the source above.

In the story

How distillation works and when it becomes an attack 13 Oct 2026