Model comparison

Qwen3-VL-2B-Instruct vs LFM2.5-VL-1.6B

At Q4_K_M, LFM2.5-VL-1.6B is the smaller download — 730,895,968 bytes against 1,107,409,952. At long context the gap widens: LFM2.5-VL-1.6B's KV cache at 32K is 9.3× smaller, which usually matters more than the difference in weights.

From the file· summed bytes, KV per layer

Side by side

Qwen3-VL-2B-InstructLFM2.5-VL-1.6B
Parameters2.1B1.6B
Architectureqwen3vllfm2
Layers2816
Native context262,144128,000
Mixture of expertsnono
Quantizations published4623
Smallest quantization0.50 GiB0.45 GiB
Q4_K_M1.03 GiB0.68 GiB
Licenceapache-2.0other

KV cache by context

the term that decides long-context viability
ContextQwen3-VL-2B-InstructLFM2.5-VL-1.6BRatio
4,0960.44 GiB0.05 GiB9.33×
8,1920.88 GiB0.09 GiB9.33×
16,3841.75 GiB0.19 GiB9.33×
32,7683.50 GiB0.38 GiB9.33×
65,5367.00 GiB0.75 GiB9.33×
131,07214.00 GiB1.50 GiB9.33×
Qwen3-VL-2B-Instruct vs LFM2.5-VL-1.6B — size, memory and hardware fit — ossmodeldb