Anton Balagaev

AB-VnimanieAI

·

https://vnimanie.ai

AI & ML interests

Constrained decoding, SGLang, vLLM, outlines, xgrammar, llguidance, inference, SDD, harnesses, agent loops, cheap KV, long context windows.

Organizations

New activity in VnimanieAI/Qwen3.8-Flash-Next-W4A16 2 months ago

Lobotomized

#4 opened 2 months ago by

Get endless model loading with 8x3090

#2 opened 2 months ago by

ways to squeeze more context?

#3 opened 2 months ago by