Model comparison

Qwen2.5-VL-7B-NSFW-Caption-V3 vs Qwythos-9B-Claude-Mythos-5-1M

At Q4_K_M, Qwen2.5-VL-7B-NSFW-Caption-V3 is the smaller download — 4,683,072,384 bytes against 11,516,776,704.

From the file· summed bytes, KV per layer

Side by side

Qwen2.5-VL-7B-NSFW-Caption-V3Qwythos-9B-Claude-Mythos-5-1M
Parameters7.6B9.4B
Architectureqwen2vlqwen35
Layers32
Native context1,048,576
Mixture of expertsnono
Quantizations published5816
Smallest quantization2.59 GiB5.38 GiB
Q4_K_M4.36 GiB10.73 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
ContextQwen2.5-VL-7B-NSFW-Caption-V3Qwythos-9B-Claude-Mythos-5-1MRatio
4,0960.13 GiB
8,1920.25 GiB
16,3840.50 GiB
32,7681.00 GiB
65,5362.00 GiB
131,0724.00 GiB