Model comparison

dolphin-2.9.3-mistral-7B-32k vs Qwen3-8B

At Q4_K_M, dolphin-2.9.3-mistral-7B-32k is the smaller download — 4,372,823,008 bytes against 5,027,783,488.

From the file· summed bytes, KV per layer

Side by side

dolphin-2.9.3-mistral-7B-32kQwen3-8B
Parameters7.2B8.2B
Architecturellamaqwen3
Layers3236
Native context32,76840,960
Mixture of expertsnono
Quantizations published3751
Smallest quantization1.50 GiB2.12 GiB
Q4_K_M4.07 GiB4.68 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
Contextdolphin-2.9.3-mistral-7B-32kQwen3-8BRatio
4,0960.50 GiB0.56 GiB1.13×
8,1921.00 GiB1.13 GiB1.13×
16,3842.00 GiB2.25 GiB1.13×
32,7684.00 GiB4.50 GiB1.13×
65,5368.00 GiB9.00 GiB1.13×
131,07216.00 GiB18.00 GiB1.13×
dolphin-2.9.3-mistral-7B-32k vs Qwen3-8B — size, memory and hardware fit — ossmodeldb