Model comparison

OLMo-2-1124-7B-Instruct 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 4,472,020,544.

From the file· summed bytes, KV per layer

Side by side

OLMo-2-1124-7B-InstructHyperCLOVAX-SEED-Text-Instruct-1.5B
Parameters7.3B1.6B
Architectureolmo2llama
Layers32
Native context4,096
Mixture of expertsnono
Quantizations published201
Smallest quantization2.49 GiB1.06 GiB
Q4_K_M4.16 GiB1.06 GiB
Licenceapache-2.0other

KV cache by context

the term that decides long-context viability
ContextOLMo-2-1124-7B-InstructHyperCLOVAX-SEED-Text-Instruct-1.5BRatio
4,0962.00 GiB
8,1924.00 GiB
16,3848.00 GiB
32,76816.00 GiB
65,53632.00 GiB
131,07264.00 GiB