TrendingRisingGemsRadarToolboxCategories

Repository

sergqwer/strata-nvfp4

Strata fork: Qwen3.8-Flash-Next 125B MoE in NVFP4 on one RTX 20-50 card (12 GB+) and 64 GB of RAM or more. Setup installs our GPTQ quants (huihui abliterated, OrcaRouter) from Hugging Face. CPU experts on AVX-512 or AVX2 (+AVX-VNNI), all experts in RAM from 92 GB, a low-RAM mode below, vision, W4A8 prompts. Windows / CUDA 13.

Open source Productivity & automation C++MIT

Stars

165

Today

+2

last 24 hours

This week

+6

Forks

17

Clone

git clone https://github.com/sergqwer/strata-nvfp4.git

Star growth

2026-10-17: +42026-10-18: +2 (today so far)2026-10-17todaymax +4/day
Stars gained per day. Tracking since 2026-10-17; the lighter bar is today so far.
#blackwell#cuda#llm-inference#moe#nvfp4#qwen#vision#windows#avx2#avx512
Created
2026-09-29
Last commit
2026-10-15
Gained since tracking
+6

Similar in Productivity & automation

Updated Sun, 18 Oct 2026 15:32:30 GMT · description from the project itself