Model comparison

Stoicism1_Llama3.1-8b-instruct vs HyperCLOVAX-SEED-Text-Instruct-1.5B

These two publish different quantization sets; the table below has the exact sizes.

From the file· summed bytes, KV per layer

Side by side

Stoicism1_Llama3.1-8b-instructHyperCLOVAX-SEED-Text-Instruct-1.5B
Parameters8.0B1.6B
Architecturellamallama
Layers32
Native context131,072
Mixture of expertsnono
Quantizations published241
Smallest quantization1.88 GiB1.06 GiB
Q4_K_M1.06 GiB
Licencecc-by-4.0other

KV cache by context

the term that decides long-context viability
ContextStoicism1_Llama3.1-8b-instructHyperCLOVAX-SEED-Text-Instruct-1.5BRatio
4,0960.50 GiB
8,1921.00 GiB
16,3842.00 GiB
32,7684.00 GiB
65,5368.00 GiB
131,07216.00 GiB