Model comparison

DeepSeek-V2-Lite-Instruct-Abliterated-15-1.2 vs Qwen3-30B-A3B-Thinking-2507

At Q4_K_M, DeepSeek-V2-Lite-Instruct-Abliterated-15-1.2 is the smaller download — 10,367,958,656 bytes against 18,556,685,824.

From the file· summed bytes, KV per layer

Side by side

DeepSeek-V2-Lite-Instruct-Abliterated-15-1.2Qwen3-30B-A3B-Thinking-2507
Parameters15.7B30.5B
Architecturedeepseek2qwen3moe
Layers48
Native context262,144
Mixture of expertsnoyes, 128 experts
Quantizations published1151
Smallest quantization6.00 GiB7.05 GiB
Q4_K_M9.66 GiB17.28 GiB
Licenceapache-2.0

KV cache by context

the term that decides long-context viability
ContextDeepSeek-V2-Lite-Instruct-Abliterated-15-1.2Qwen3-30B-A3B-Thinking-2507Ratio
4,0960.38 GiB
8,1920.75 GiB
16,3841.50 GiB
32,7683.00 GiB
65,5366.00 GiB
131,07212.00 GiB