Model comparison

openchat-3.5-0106 vs Qwen3-VL-8B-Instruct-abliterated-v1

At Q4_K_M, openchat-3.5-0106 is the smaller download — 4,368,450,656 bytes against 5,027,785,792.

From the file· summed bytes, KV per layer

Side by side

openchat-3.5-0106Qwen3-VL-8B-Instruct-abliterated-v1
Parameters7.2B8.8B
Architecturellamaqwen3vl
Layers32
Native context8,192
Mixture of expertsnono
Quantizations published4548
Smallest quantization1.50 GiB1.97 GiB
Q4_K_M4.07 GiB4.68 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
Contextopenchat-3.5-0106Qwen3-VL-8B-Instruct-abliterated-v1Ratio
4,096
8,192
16,384
32,768
65,536
131,072