---
id: 7631
title: "GLM and DeepSeek Are Catching Frontier Models at Finding Vulnerabilities"
source_domain: hacktron.ai
source_url: "https://www.hacktron.ai/blog/open-weight-models-vs-frontier-models"
published_at: "2026-07-08T00:00:00Z"
tags: ["vulnerability-discovery", "glm", "llm-benchmark", "deepseek", "open-weight-models", "fider"]
phying_url: "https://phying.com/item/7631"
lang: en
---

# GLM and DeepSeek Are Catching Frontier Models at Finding Vulnerabilities

## Summary

Benchmarking GLM, DeepSeek, Kimi, and MiniMax against GPT, Claude, Gemini, and Opus on 15 real Fider vulnerabilities: frontier models still lead, but open-weight models are closing the gap fast.

## Intelligence Metadata

- **Sources:** hacktron.ai
- **Published:** 2026-07-08T00:00:00Z
- **Original:** https://www.hacktron.ai/blog/open-weight-models-vs-frontier-models
- **Phying News:** https://phying.com/item/7631