Model comparison

Olmo-3-7B-Instruct vs Qwen3-VL-8B-Instruct-abliterated-v1

At Q4_K_M, Olmo-3-7B-Instruct is the smaller download — 4,471,601,152 bytes against 5,027,785,792.

From the file· summed bytes, KV per layer

Side by side

Olmo-3-7B-InstructQwen3-VL-8B-Instruct-abliterated-v1
Parameters7.3B8.8B
Architectureolmo2qwen3vl
Layers32
Native context65,536
Mixture of expertsnono
Quantizations published4648
Smallest quantization1.81 GiB1.97 GiB
Q4_K_M4.16 GiB4.68 GiB
Licenceapache-2.0apache-2.0

KV cache by context

the term that decides long-context viability
ContextOlmo-3-7B-InstructQwen3-VL-8B-Instruct-abliterated-v1Ratio
4,0962.00 GiB
8,1922.69 GiB
16,3843.69 GiB
32,7685.69 GiB
65,5369.69 GiB
131,07217.69 GiB