Model comparison

Qwen3-Coder-30B-A3B-Instruct vs solar-pro-preview-instruct

At Q4_K_M, solar-pro-preview-instruct is the smaller download — 13,310,219,072 bytes against 18,556,689,568.

From the file· summed bytes, KV per layer

Side by side

Qwen3-Coder-30B-A3B-Instructsolar-pro-preview-instruct
Parameters30.5B22.1B
Architectureqwen3moellama
Layers4864
Native context262,1444,096
Mixture of expertsyes, 128 expertsno
Quantizations published4615
Smallest quantization7.46 GiB4.46 GiB
Q4_K_M17.28 GiB12.40 GiB
Licenceapache-2.0

KV cache by context

the term that decides long-context viability
ContextQwen3-Coder-30B-A3B-Instructsolar-pro-preview-instructRatio
4,0960.38 GiB
8,1920.75 GiB
16,3841.50 GiB
32,7683.00 GiB
65,5366.00 GiB
131,07212.00 GiB