Skip to content
X · @togethercompute · X / Twitter

We analyzed DeepSeek V4 Flash and GPT-5.6 Luna on DeepSWE. A DeepSeek-first cascade with test-suite verification solved MORE tasks than Luna alone at …

DeepSeek GPT

We analyzed DeepSeek V4 Flash and GPT-5.6 Luna on DeepSWE.A DeepSeek-first cascade with test-suite verification solved MORE tasks than Luna alone at 37% lower cost per task.

Read original