Model comparison

nemotron-3.5-asr-streaming-0.6b vs GLM-ASR-Nano-2512

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

From the file· summed bytes, KV per layer

Side by side

nemotron-3.5-asr-streaming-0.6bGLM-ASR-Nano-2512
Parameters638M2.3B
Architectureparakeetglmasr
Layers28
Native context8,192
Mixture of expertsnono
Quantizations published85
Smallest quantization0.38 GiB0.91 GiB
Q4_K_M0.46 GiB
Licenceothermit

KV cache by context

the term that decides long-context viability
Contextnemotron-3.5-asr-streaming-0.6bGLM-ASR-Nano-2512Ratio
4,0960.22 GiB
8,1920.44 GiB
16,3840.88 GiB
32,7681.75 GiB
65,5363.50 GiB
131,0727.00 GiB