Model comparison

Qwen3-Coder-Next-DFlash vs ced-base

These two publish different quantization sets; the table below has the exact sizes.

From the file· summed bytes, KV per layer

Side by side

Qwen3-Coder-Next-DFlashced-base
Parameters474M86M
Architecturedflashced
Layers8
Native context262,144
Mixture of expertsnono
Quantizations published73
Smallest quantization0.25 GiB0.12 GiB
Q4_K_M0.27 GiB
Licenceapache-2.0

KV cache by context

the term that decides long-context viability
ContextQwen3-Coder-Next-DFlashced-baseRatio
4,0960.06 GiB
8,1920.13 GiB
16,3840.25 GiB
32,7680.50 GiB
65,5361.00 GiB
131,0722.00 GiB