DeepSeek V4 Flash Vision Abliterated vs GLM 5.3 Flash EXL3 running both on 2 sparks 👇
The question: What is the highest leverage thing humanity still doesn’t know that it could know in the next few decades? The question whose answer would most compress our ignorance about
DeepSeek V4.1 Flash over the API is *really* fast!
🤯 Some details: • new architecture + native multimodality • 1M context window • same price as v4 Flash For me it seems to be (over)thinking a lot though.
The beta version of DeepSeek 4.1 Flash has recorded very fast performance in tests.
See the output speed test in this video: 👇
Deepseek-v4.1-flash I'm really curious to understand the architecture behind the scenes.
🤔 Model is really fast as you can see in the video! I've got an average decoding speed of around 350 t/s. BTW it is clearly experimental and: - currently, the model is under heavy load and
You have reached the end of the archive
All of deepseek