Model comparison

EXAONE-3.5-2.4B-Instruct vs Qwen3-VL-8B-Instruct-abliterated-v1

At Q4_K_M, EXAONE-3.5-2.4B-Instruct is the smaller download — 1,644,918,560 bytes against 5,027,785,792.

From the file· summed bytes, KV per layer

Side by side

EXAONE-3.5-2.4B-InstructQwen3-VL-8B-Instruct-abliterated-v1
Parameters2.4B8.8B
Architectureexaoneqwen3vl
Layers
Native context32,768
Mixture of expertsnono
Quantizations published2648
Smallest quantization0.95 GiB1.97 GiB
Q4_K_M1.53 GiB4.68 GiB
Licenceotherapache-2.0

KV cache by context

the term that decides long-context viability
ContextEXAONE-3.5-2.4B-InstructQwen3-VL-8B-Instruct-abliterated-v1Ratio
4,096
8,192
16,384
32,768
65,536
131,072