image/png

V2.4 of Dungeonmaster (the most creative model so far in the Dungeonmaster series), I decided to move away from the R1 base here, because I feel it the pros dont necessarily outweigh the cons. For the V2.X series I decided to go for a custom uncensored base.

Shoutout to Kromeurus for helping me with this novel stepped merge methodology and for Geechan for testing and supplying the recommended sampler settings:

1 Temp and 0.01 Min P.

Dungeonmaster is meant to be specifically for creative roleplays with stakes and consequences using the following curated models:

Support on KO-FI <3

Downloads last month
97
Safetensors
Model size
70.6B params
Tensor type
BF16
·
Inference Providers NEW
The selected billing account doesn't have any compatible Inference Provider enabled for this model. Settings

Model tree for Tarek07/Dungeonmaster-V2.4-Expanded-LLaMa-70B