PrototypeCommon
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…
- Owner
- sergqwer (personal account)
- Stars
- Stars: 217
- Forks
- Forks: 23
- Language
- C++
- License
- MIT
- Created
- 2026
- Topics
- blackwellcudallm-inferencemoenvfp4qwenvisionwindowsavx2avx512
- Tier
- Prototype (100 to 249 ★) What this tier means
Nearby in C++
All C++ repos- ouwou/acheronStars: 217C++Prototype
- HackMan3D/HackMan3D-Orbit-ControllerStars: 217C++Prototype
- CreoleVR/QuestLHSyncStars: 217C++Prototype
- neubig/kyteaStars: 217C++Prototype
- tomahawk-player/tomahawk-resolversStars: 217C++Prototype
- pypt/fervorStars: 217C++Prototype