-OZONENEWS-

Independent · Verified · In-Depth

Kimi K26 Browsecomp Agent Benchmark

1 article tagged with “Kimi K26 Browsecomp Agent Benchmark”. Verified news, source-cited reporting, and in-depth coverage from OzoneNews.

1 article

Kimi K2.6 | Moonshot AI Open-Weights Model Beats GPT-5.4
News

Kimi K2.6 | Moonshot AI Open-Weights Model Beats GPT-5.4

Moonshot AI releases Kimi K2.6, a 1-trillion parameter open-weights model that outscores GPT-5.4 and Claude Opus 4.6 on SWE-Bench Pro, BrowseComp, and Humanity's Last Exam. API priced at $0.95 per million tokens.

Alfasa · Apr 21, 2026