Model comparison

DS-R1-Qwen3-8B-ArliAI-RpR-v4-Small vs Qwen3-4B

These two publish different quantization sets; the table below has the exact sizes.

From the file· summed bytes, KV per layer

Side by side

DS-R1-Qwen3-8B-ArliAI-RpR-v4-SmallQwen3-4B
Parameters8.2B4.0B
Architectureqwen3qwen3
Layers3636
Native context131,07240,960
Mixture of expertsnono
Quantizations published2629
Smallest quantization1.97 GiB1.01 GiB
Q4_K_M2.33 GiB
Licenceapache-2.0

KV cache by context

the term that decides long-context viability
ContextDS-R1-Qwen3-8B-ArliAI-RpR-v4-SmallQwen3-4BRatio
4,0960.56 GiB0.56 GiB1.00×
8,1921.13 GiB1.13 GiB1.00×
16,3842.25 GiB2.25 GiB1.00×
32,7684.50 GiB4.50 GiB1.00×
65,5369.00 GiB9.00 GiB1.00×
131,07218.00 GiB18.00 GiB1.00×