
Intelligence is becoming abundant. Access to it should be, too. Today, we’re introducing Alpha. We’re building the inference infrastructure for a world where open intelligence becomes the default. Starting with Europe. alpha.sh/blog/introducing-al…

The performance layer for sovereign open intelligence. We run open-weight models 2× faster and 30% cheaper on US & EU capacity reserved for you.

Intelligence is becoming abundant. Access to it should be, too. Today, we’re introducing Alpha. We’re building the inference infrastructure for a world where open intelligence becomes the default. Starting with Europe. alpha.sh/blog/introducing-al…

open-weight models closed the quality gap. the new bottleneck is serving: cost per token, latency under load, and where your data is processed. Alpha has joined NVIDIA Inception to scale that layer. alpha.sh/blog/alpha-joins-nv…

alpha is now live on @useRouterPlus.
@alpha_dot_sh is now live on routerplus.

DeepSeek V4.1 Flash 1017 is live on Alpha.sh.
We found DeepSeek V4.1 Flash 1017 beats GPT 6 Sol on our multi-step offensive cybersecurity task. Sol was tested via the Daybreak trusted access program. We have been building a cybersecurity benchmark based on simulated corporate networks. This is a valuable indicator for the cyber-risk a model poses if in the hands of a bad actor. We plan to make the full evaluation public in the coming weeks but wanted to share our first immediate finding here.

Welcoming @BlockRunAI on alpha.sh.
DeepSeek V4.1 Flash is now live on BlockRun. One model. Multiple ways to access it: Base, Solana, Arc, or API key. Pricing per 1M tokens → Input: $0.14 → Output: $0.28 → Cached input: $0.028 → $0.006 Terminal-Bench 4.0 improved from 31.2 to 42.3, while DeepSWE rose from 74.2 to 77.0. Try it on BlockRun 👇 blockrun.ai/models/deepseek-…


Alpha.sh is live on Liquid Inference.
The Liquid Inference router network is growing quickly. We now have 28 verified companies serving >1000 models, offering capacity dynamically and updating quotes in real-time. Average time for providers to onboard and integrate models is 3 hours, full verification within a day.

Hello world.