Model comparison

Qwen-STEM-Specialist-7B 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 4,683,074,560.

From the file· summed bytes, KV per layer

Side by side

Qwen-STEM-Specialist-7BHyperCLOVAX-SEED-Text-Instruct-1.5B
Parameters7.6B1.6B
Architectureqwen2llama
Layers28
Native context32,768
Mixture of expertsnono
Quantizations published361
Smallest quantization1.77 GiB1.06 GiB
Q4_K_M4.36 GiB1.06 GiB
Licenceother

KV cache by context

the term that decides long-context viability
ContextQwen-STEM-Specialist-7BHyperCLOVAX-SEED-Text-Instruct-1.5BRatio
4,0960.22 GiB
8,1920.44 GiB
16,3840.88 GiB
32,7681.75 GiB
65,5363.50 GiB
131,0727.00 GiB
Qwen-STEM-Specialist-7B vs HyperCLOVAX-SEED-Text-Instruct-1.5B — size, memory and hardware fit — ossmodeldb