Model comparison

gaon-1.7b-v2-translate vs HyperCLOVAX-SEED-Text-Instruct-1.5B

At Q4_K_M, gaon-1.7b-v2-translate is the smaller download — 1,107,409,248 bytes against 1,133,974,368.

From the file· summed bytes, KV per layer

Side by side

gaon-1.7b-v2-translateHyperCLOVAX-SEED-Text-Instruct-1.5B
Parameters1.7B1.6B
Architectureqwen3llama
Layers28
Native context2,048
Mixture of expertsnono
Quantizations published361
Smallest quantization0.48 GiB1.06 GiB
Q4_K_M1.03 GiB1.06 GiB
Licenceapache-2.0other

KV cache by context

the term that decides long-context viability
Contextgaon-1.7b-v2-translateHyperCLOVAX-SEED-Text-Instruct-1.5BRatio
4,0960.44 GiB
8,1920.88 GiB
16,3841.75 GiB
32,7683.50 GiB
65,5367.00 GiB
131,07214.00 GiB