Model comparison

gpt2 vs ced-base

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

From the file· summed bytes, KV per layer

Side by side

gpt2ced-base
Parameters137M86M
Architecturegpt2ced
Layers12
Native context
Mixture of expertsnono
Quantizations published493
Smallest quantization0.06 GiB0.12 GiB
Q4_K_M0.11 GiB
Licenceapache-2.0

KV cache by context

the term that decides long-context viability
Contextgpt2ced-baseRatio
4,096
8,192
16,384
32,768
65,536
131,072
gpt2 vs ced-base — size, memory and hardware fit — ossmodeldb