Deepseek-v4.1-flash is just⚡️
I'm not getting used to this speed. Getting ~250 tok/s average per stream even now. I cannot really imagine how fast next deepseek-v4.1-flash-fast is going to be. Might hit 500 tok/s
Introducing askr Anon. Private group chats, private AI, when you need them.
Askr remembers what's worth keeping. Anon is for the chats you don't want kept. No account. Open and start. You get a random guest name and nothing else: no email, no password,
You thought I was joking 🤓
Here's Deepseek-v4.1-Flash running on CPU + DDR4 with just a single RTX 3090 Wow MiaAI_lab great quant (:
How fast do tokens move to a new model?
Share of tokens on the new model vs. its predecessor, first 72 hours on Requesty: - GPT-6 Sol = 49% - Claude Fable 5.1 = 43% - GPT-6 Luna = 35% - Claude Opus 5.5 = 23% - DeepSeek V4.1 Flash = 23% - Claude Sonnet 5.5 = 5%
You have reached the end of the archive
All of deepseek