Model comparison

antares-1b vs Qwen3-1.7B

At Q4_K_M, Qwen3-1.7B is the smaller download — 1,107,409,472 bytes against 1,139,251,136.

From the file· summed bytes, KV per layer

Side by side

antares-1bQwen3-1.7B
Parameters1.8B2.0B
Architecturegraniteqwen3
Layers28
Native context40,960
Mixture of expertsnono
Quantizations published848
Smallest quantization0.72 GiB0.50 GiB
Q4_K_M1.06 GiB1.03 GiB
Licenceapache-2.0

KV cache by context

the term that decides long-context viability
Contextantares-1bQwen3-1.7BRatio
4,0960.44 GiB
8,1920.88 GiB
16,3841.75 GiB
32,7683.50 GiB
65,5367.00 GiB
131,07214.00 GiB