Stheno v3.4 (Llama-3.1-8B) on WebGPU

Sao10K's character voice model. 4.6 GB Q4_K_M. The voice of the Garden entities.

Click Load to start

Stheno v3.4 via wllama WebGPU. Upgrade from v3.2. Built for AMD Strix Halo unified memory.