cd /news/large-language-models/kimi-k2-6-reasoning-thinking-loop · home › topics › large-language-models › article
[ARTICLE · art-147003] src=kagifeedback.org ↗ pub= topic=large-language-models verified=true sentiment=↓ negative

Kimi K2.6 Reasoning thinking loop

A user reported that Kimi K2.6 entered a reasoning thinking loop, repeating the same words roughly 1000 times, and that the stop button failed to cancel the request, which ran to the model's output limit and cost 0.21€. The user said the cancel button should terminate the request rather than letting it run to the output limit, and that the model should automatically terminate after detecting such a looping response.

read1 min views1 publishedOct 7, 2026

I had this looping request (idk if the image above works) and tried to cancel it with the stop button, but that didn't work. Luckily it's just 0.21€, but still very unnecessary that it had to run until the very end of the model output limits instead of accepting my cancellation request or automatically terminating after detecting such a looping response (after 1000x same words).

I expect the cancel button to cancel the request instead of running till the output limit.

── more in #large-language-models 4 stories · sorted by recency
── more on @kimi k2.6 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/kimi-k2-6-reasoning-…] indexed:0 read:1min 2026-10-07 · —