Anthropic says Zhipu GLM-5.3 exploit capabilities approach Claude
What happened
On 30 September 2026, Anthropic published an assessment finding that Zhipu's open-weight GLM-5.3 model approached Claude Mythos Preview in exploit development capabilities. That day, the Anthropic Frontier Red Team released details: across 100 random tasks in its internal Binary Exploitation benchmark, GLM-5.3 achieved full control-flow hijacking in 4% of trials, versus 6% for Claude Mythos Preview. Earlier models such as Claude Opus 4.6 and GLM-5.2 had not succeeded on these tasks, and the report said a meaningful capability threshold had been crossed. Initial reports mentioned the close capabilities without methods, scores or comparisons; subsequent reporting supplied the 4% and 6% figures and benchmark details. No response from Zhipu has been reported.
Written by AI from the reports · updated 23 h ago
Timeline
Follow the reports to see the event from each side.
- The DecoderSelectedAnthropic says Zhipu's open-weight GLM-5.3 nearly matches Claude Mythos Preview at building exploits
Anthropic published an evaluation saying Zhipu's open-weight GLM-5.3 model approaches its Claude Mythos Preview in exploit development.
- Simon WillisonQuoting Anthropic Frontier Red Team
Anthropic Frontier Red Team evaluated several models on 100 random tasks from its internal Binary Exploitation benchmark. GLM-5.3 achieved full control-flow hijacking in 4% of trials, versus 6% for Claude Mythos Preview. The report says earlier models such as Claude Opus 4.6 and GLM-5.2 had not succeeded on these tasks, suggesting a meaningful capability threshold has been crossed.
Heat of this event
Current heat 8·Peak in the comparable range 15（Sep 30 23:00）·Change over 24 hours in the comparable range –
The trend compares only the same accounts observed without a break, so its range can be narrower than the current heat count. Move the pointer or tap the chart to see each hour; with a keyboard, use the left and right arrow keys.