Model comparison

Qwen3-30B-A3B-Gemini-Pro-High-Reasoning-2507-ABLITERATED-UNCENSORED vs Qwen3-30B-A3B-Thinking-2507

At Q4_K_M, Qwen3-30B-A3B-Thinking-2507 is the smaller download — 18,556,685,824 bytes against 18,556,687,648.

From the file· summed bytes, KV per layer

Side by side

Qwen3-30B-A3B-Gemini-Pro-High-Reasoning-2507-ABLITERATED-UNCENSOREDQwen3-30B-A3B-Thinking-2507
Parameters30.5B30.5B
Architectureqwen3moeqwen3moe
Layers4848
Native context262,144262,144
Mixture of expertsyes, 128 expertsyes, 128 experts
Quantizations published5751
Smallest quantization5.98 GiB7.05 GiB
Q4_K_M17.28 GiB17.28 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
ContextQwen3-30B-A3B-Gemini-Pro-High-Reasoning-2507-ABLITERATED-UNCENSOREDQwen3-30B-A3B-Thinking-2507Ratio
4,0960.38 GiB0.38 GiB1.00×
8,1920.75 GiB0.75 GiB1.00×
16,3841.50 GiB1.50 GiB1.00×
32,7683.00 GiB3.00 GiB1.00×
65,5366.00 GiB6.00 GiB1.00×
131,07212.00 GiB12.00 GiB1.00×
Qwen3-30B-A3B-Gemini-Pro-High-Reasoning-2507-ABLITERATED-UNCENSORED vs Qwen3-30B-A3B-Thinking-2507 — size, memory and hardware fit — ossmodeldb