BridgeMind(@bridgemindai):DeepSeek v4 Flash is actually pretty good. Is it as good as the benchmarks claim? No. It is benchmaxxed like everything else shipping right now. But it is a 284B parameter model. At that size it has no business performing this well. It beats GPT 5.6 Luna in real use. A model a fraction of the size, from a lab a fraction of the budget. DeepSeek keeps doing more with less than anyone in the world. Very impressed.

BridgeMind

@bridgemindai

The home of the vibe coding movement. Founded by @matthewmillerai. 98k+ on YouTube. Building to $1M in public.

加入 March 2024

8 正在關注 48.9K 粉絲

BridgeMind@bridgemindai

2026.07.31 18:44

DeepSeek v4 Flash is actually pretty good. Is it as good as the benchmarks claim? No. It is benchmaxxed like everything else shipping right now. But it is a 284B parameter model. At that size it has no business performing this well. It beats GPT 5.6 Luna in real use. A model a fraction of the size, from a lab a fraction of the budget. DeepSeek keeps doing more with less than anyone in the world. Very impressed.