The most efficient AI inference stack for async workloads

London, UK
Inkling by @thinkymachines is live on sference from day 0. 975B MoE · 41B active · 256k context Hosted on European GPUs. Try it via the sference API: sference.com
3
954
Who wants access to moonshotai/Kimi-K2.7-Code hosted on European servers by a European company? It's free for now. And you can use it in claude code. Comment "euromaxxing" to get access.
7
5
14
587