.@deepseek_ai-V4-Pro-0813 is live on DigitalOcean Serverless Inference and Inference Router. do.co/3U24Cug
1.6T params / 49B active, 1M-token context, AA Intelligence Index of 53 at max reasoning effort.
Were closing out Ray Summit + the vLLM Conference with our happy hour, hosted with @nvidia & @inferact.
Lightning talks, live demos, cold drinks, and a room full of builders who are optimizing intelligence per dollar.
RSVP today. do.co/4g0oAhE
Most people wait months for the latest GPUs, like @AMD Instinct MI350 or 355s.
We spun one up, installed @vllm_project and @Alibaba_Qwen 2.5-72B, and had it explain quantum physics in detail. Total time: under 2 minutes.
Spot GPU Droplets are now in Public Preview. Show more
.@amit's repo-lens takes @Alibaba_Qwen 3.8's full context window and turns it into an honest, cited read of what a dependency actually does. Built on Serverless Inference.
Alibaba dropped Qwen 3.8 last week, and it's now on @digitalocean Serverless Inference.
The thing that got my attention is the context window, hundreds of thousands of tokens in one pass. Basically a whole codebase.
So I threw an entire repo at it.
Now available: @Alibaba_Qwen 3.8-2.4T-A95B from @alibaba_cloud on DigitalOcean Serverless Inference via NVIDIA HGX B300 GPUs. 1M context, built for long-horizon coding. do.co/45VBAyZ
One API, usage-based pricing, no infra to manage.