Anyone has tested Qwen3.8-Flash-Next with Strata in Sillytavern?
Strata with Qwen3.8-Flash-Next is ridiculously fast on my 5080 with 16gb VRAM and 64GB System Ram. I tried to use it with Sillytavern via OpenAI compatible endpoint with Chat Completion, but the answers I get are not great, if any. Anyone has managed to make this work?