Model comparison

Darwin-4B-Chimera vs Qwen3-VL-8B-Instruct-abliterated-v1

At Q4_K_M, Darwin-4B-Chimera is the smaller download — 2,497,281,088 bytes against 5,027,785,792.

From the file· summed bytes, KV per layer

Side by side

Darwin-4B-ChimeraQwen3-VL-8B-Instruct-abliterated-v1
Parameters4.0B8.8B
Architectureqwen3qwen3vl
Layers36
Native context40,960
Mixture of expertsnono
Quantizations published5948
Smallest quantization0.98 GiB1.97 GiB
Q4_K_M2.33 GiB4.68 GiB
Licencegemmaapache-2.0

KV cache by context

the term that decides long-context viability
ContextDarwin-4B-ChimeraQwen3-VL-8B-Instruct-abliterated-v1Ratio
4,0960.56 GiB
8,1920.71 GiB
16,3840.90 GiB
32,7681.28 GiB
65,5362.03 GiB
131,0723.53 GiB