Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
Llama_3.2_3b_Kermes_v2.1-iMat-CQ-GGUF – AI Model by NexesQuants | AlphaNeural AI
You can deploy this model and start earning money today!
NexesQuants
/
Llama_3.2_3b_Kermes_v2.1-iMat-CQ-GGUF
like
0
Nexesenex/Llama_3.2_3b_Kermes_v2.1
quantized
conversational
endpoints_compatible
gguf
template
llama3.2
us
Views
No views yet
Model card
Files and Versions
Community
API
Deploy
GGUF static quantizations (Thanks!) :
https://huggingface.co/mradermacher/Llama_3.2_3b_Kermes_v2.1-GGUF
GGUF custom quantizations:
Arc-C is around 47-48.
Arc-E around 70-71.
Perplexity at 512 ctx, wikitext english :
BF16 : 9.11
Best :
Q8_0 : 9.11
Recommended :
Q6_K : 9.17
Q5_K : 9.18
Best compromises :
Q4_K : 9.24
IQ4_S : 9.26
Good compromises :
IQ4_XS : 9.38
Q3_K : 9.52
IQ3_L : 9.58
Usable still :
IQ3_XS : 9.97
For the poor :
Q2_K : 11.06
IQ2_XL : 11.56
