How I select models ? First, I bench them, to trim the overfit & dumbified models. Then, I test them. The smartest to my taste end-up here.
Nexes the Elder
Nexesenex
AI & ML interests
Maintaining KoboldCPP fork Croco.Cpp.
Looting PRs here and there.
Merging since the 11/02/2025, happily so. Stay tuned!
Making quants often requiring IK_Llama.cpp or Croco.Cpp to run.
Sharing my favorite models quants (recently 70b-123b for bi and tri GPU rigs).
Barking at many trees.
Thanks to Mradermacher and Bartowski for their imatrixes, I'm using them extensively.
Recent Activity
updated
a model
less than a minute ago
Nexesenex/Llama_3.2_1b_RandomLego_RP_R1_0.1
updated
a model
1 minute ago
Nexesenex/Llama_3.2_1b_OpenTree_R1_0.1
liked
a model
2 minutes ago
Nexesenex/Llama_3.2_1b_RandomLego_RP_R1_0.1
Organizations
Collections
2
models
179
Nexesenex/Llama_3.2_1b_RandomLego_RP_R1_0.1
Text Generation
•
Updated
•
1
Nexesenex/Llama_3.2_1b_OpenTree_R1_0.1
Text Generation
•
Updated
•
1
Nexesenex/Llama_3.2_1b_Dolto_0.1
Text Generation
•
Updated
•
1
Nexesenex/Llama_3.2_3b_Kermes_v2.1
Text Generation
•
Updated
•
33
•
2
Nexesenex/Llama_3.2_3b_Kermes_v2
Text Generation
•
Updated
•
84
•
2
Nexesenex/Llama_3.2_3b_BlueKermes_0.39-GGUF
Updated
•
110
Nexesenex/Llama_3.2_3b_PlayMess_0.1
Text Generation
•
Updated
•
69
•
2
Nexesenex/Llama_3.2_3b_KermesRed_V3.4-GGUF
Updated
•
107
Nexesenex/Llama_3.2_3b_KermesPink_V3.6-GGUF
Updated
•
87
Nexesenex/Llama_3.x_70b_BetsyTeaz_0.1-iMat-CG-GGUF
Updated
•
51
datasets
None public yet