@estebankiwiDeepSeek v4 Flash is genuinely impressive. It's a 284B parameter model that outdoes GPT 5.6 Luna in real-world use, which feels unfair at that size. Sure, it's benchmaxxed like everything else, but the performance-to-size ratio is wild. DeepSeek keeps punching way above their weight.
View original postPhoto by @estebankiwi· Jul 31, 2026· DeepSeek
The image is a bar chart titled "Artificial Analysis Coding Index". It displays various AI models and their performance scores, represented by colored bars. The mood is informative and data-driven, with a clean, modern aesthetic. A notable detail is the red outline around the blue bar representing "DeepSeek V4 Flash 0731 (max)", highlighting it as a point of interest. The chart includes logos for each AI model and their names are listed below the bars.
@estebankiwiDeepSeek v4 Flash is genuinely impressive. It's a 284B parameter model that outdoes GPT 5.6 Luna in real-world use, which feels unfair at that size. Sure, it's benchmaxxed like everything else, but the performance-to-size ratio is wild. DeepSeek keeps punching way above their weight.
View original postPhoto by @estebankiwi· Jul 31, 2026· DeepSeek
The image is a bar chart titled "Artificial Analysis Coding Index". It displays various AI models and their performance scores, represented by colored bars. The mood is informative and data-driven, with a clean, modern aesthetic. A notable detail is the red outline around the blue bar representing "DeepSeek V4 Flash 0731 (max)", highlighting it as a point of interest. The chart includes logos for each AI model and their names are listed below the bars.
