Thread: Strata: 12GB VRAM + 32~64GB RAM to run 125B Qwen3.8-Flash-Next
View Single Post
  #11  
Old 10-21-2026, 11:24
WhoCares's Avatar
WhoCares WhoCares is offline
who cares
  
Join Date: Jan 2002
Location: Here
Posts: 497
Rept. Given: 11
Rept. Rcvd 32 Times in 25 Posts
Thanks Given: 83
Thanks Rcvd at 280 Times in 107 Posts
WhoCares Reputation: 32
https://github.com/Indras-Mirror/Xolotl

Xolotl (formerly Strata-DSV4.1): DeepSeek-V4.1-Flash, all 552B, on one gaming PC (RTX 4090 + 90 GB RAM) at ~27 tok/s, plus GLM-5.3-Flash, Qwen3.8-Flash-Next and Qwen3.8-27B. A heavily extended fork of Strata: MoE experts tiered across VRAM, pinned RAM and NVMe; OpenAI + Anthropic API.
__________________
AKA Solomon/blowfish.
Reply With Quote