We burned 11.7bn tokens to find the best cyber AI model | GLM5.3 and DeepSeek are now frontier

https://www.aikido.dev/blog/ai-model-benchmarks-aug-21-2026

https://x.com/home/status/2106906890357936227

https://x.com/home/status/2106927516611125447

https://x.com/home/status/2107211540134826379