Model comparison

MARTHA-LXVII.8B_QWEN-3.5-3.6_prune_9b-3.6_base vs HyperCLOVAX-SEED-Text-Instruct-1.5B

At Q4_K_M, HyperCLOVAX-SEED-Text-Instruct-1.5B is the smaller download — 1,133,974,368 bytes against 5,106,435,872.

From the file· summed bytes, KV per layer

Side by side

MARTHA-LXVII.8B_QWEN-3.5-3.6_prune_9b-3.6_baseHyperCLOVAX-SEED-Text-Instruct-1.5B
Parameters8.1B1.6B
Architectureqwen35llama
Layers28
Native context262,144
Mixture of expertsnono
Quantizations published361
Smallest quantization2.35 GiB1.06 GiB
Q4_K_M4.76 GiB1.06 GiB
Licenceapache-2.0other

KV cache by context

the term that decides long-context viability
ContextMARTHA-LXVII.8B_QWEN-3.5-3.6_prune_9b-3.6_baseHyperCLOVAX-SEED-Text-Instruct-1.5BRatio
4,0960.11 GiB
8,1920.22 GiB
16,3840.44 GiB
32,7680.88 GiB
65,5361.75 GiB
131,0723.50 GiB