Is DeepSeek-V4-Flash or MiniMax-M2 smaller?

Both publish quantizations; compare the table for the exact sizes.

Which uses less memory at long context?

DeepSeek-V4-Flash — its KV cache at 32K context is 123.0x smaller, because of how its attention layers cache.

Model comparison

DeepSeek-V4-Flash vs MiniMax-M2

These two publish different quantization sets; the table below has the exact sizes. At long context the gap widens: DeepSeek-V4-Flash's KV cache at 32K is 123.0× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	DeepSeek-V4-Flash	MiniMax-M2
Parameters	291B	229B
Architecture	deepseek4	minimax-m2
Layers	43	62
Native context	1,048,576	196,608
Mixture of experts	yes, 256 experts	yes, 256 experts
Quantizations published	12	48
Smallest quantization	76.87 GiB	43.78 GiB
Q4_K_M	—	128.84 GiB
Licence	mit	mit

KV cache by context

the term that decides long-context viability

Context	DeepSeek-V4-Flash	MiniMax-M2	Ratio
4,096	0.06 GiB	0.97 GiB	15.38×
8,192	0.06 GiB	1.94 GiB	30.76×
16,384	0.06 GiB	3.88 GiB	61.52×
32,768	0.06 GiB	7.75 GiB	123.04×
65,536	0.06 GiB	15.50 GiB	246.08×
131,072	0.06 GiB	31.00 GiB	492.16×