edtorre/ gemma4-31b-hermes

39 2 weeks ago

Gemma 4 31B (Q4_0) optimized for Hermes Agent — 64K context, 8192 max tokens, flash attention + Q8 KV cache. Fits 24 GB VRAM with optimizations.

vision tools thinking

1 model

gemma4-31b-hermes:latest
615f1a4c56f6 • 19GB • 256K context window •
Text, Image input • 2 weeks ago