Kimi K3 finds 16 new vulnerabilities and beats GLM-5.2 on an exploit benchmark
zephyr_z9 · x · 2026-07-28
Kimi K3’s cyber report shows 16 new bugs and better exploit performance than GLM-5.2
The Kimi K3 tech report says the model found 16 previously unknown vulnerabilities across six projects, including two Linux kernel bugs: a remotely triggerable DoS and a local privilege-escalation issue.
On an exploit-development suite, Kimi K3 solved 14/36 tasks (38.9%) versus 8/36 (22.2%) for GLM-5.2. But the gains were uneven: 10 of Kimi’s 14 wins came from the user-space track, and both models failed about three quarters of the kernel track. The report also notes that the remaining misses cluster around incomplete exploit chains, poor strategy selection under mitigations, unproductive debugging loops, and insufficient final verification.
More from Models
- MAI-Cyber-1-Flash is built to cover 90% of CyberGym patching work — mustafasuleyman · 2026-07-28
- Gemini video generation adds words and blocks some harmless prompts — Individual-Cookie615 · 2026-07-28
- Polymarket now prices a 42% chance of a new Claude Sonnet by next month — Polymarket · 2026-07-28
- Kimi K3 paper details SiTU-GLU and quantile balancing for 896-expert MoE — KyeGomezB · 2026-07-28
- vLLM Collaborates with DigitalOcean to Host Kimi K3 Model — vllm_project · 2026-07-28
- Kimi K3 lands on ChatLLM with U.S. hosting and an open-source fine-tune — bindureddy · 2026-07-28