Is CodeLlama-70b-Instruct-hf or Qwen3.6-27B smaller?

Qwen3.6-27B is smaller at Q4_K_M: 15.41 GiB against 38.58 GiB.

Which uses less memory at long context?

Qwen3.6-27B — its KV cache at 32K context is 5.0x smaller, because of how its attention layers cache.

Model comparison

CodeLlama-70b-Instruct-hf vs Qwen3.6-27B

At Q4_K_M, Qwen3.6-27B is the smaller download — 16,547,397,888 bytes against 41,423,092,736. At long context the gap widens: Qwen3.6-27B's KV cache at 32K is 5.0× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

	CodeLlama-70b-Instruct-hf	Qwen3.6-27B
Parameters	69.0B	27.8B
Architecture	llama	qwen35
Layers	80	64
Native context	4,096	262,144
Mixture of experts	no	no
Quantizations published	40	40
Smallest quantization	13.54 GiB	8.74 GiB
Q4_K_M	38.58 GiB	15.41 GiB
Licence	llama2	apache-2.0

KV cache by context

the term that decides long-context viability

Context	CodeLlama-70b-Instruct-hf	Qwen3.6-27B	Ratio
4,096	1.25 GiB	0.25 GiB	5.00×
8,192	2.50 GiB	0.50 GiB	5.00×
16,384	5.00 GiB	1.00 GiB	5.00×
32,768	10.00 GiB	2.00 GiB	5.00×
65,536	20.00 GiB	4.00 GiB	5.00×
131,072	40.00 GiB	8.00 GiB	5.00×