Skip to content

Can A100 80G run Nemotron 3.5 Super VL 120B-A12B?

Tight: GGUF Q4_0 needs 75.4 GB at 4K, batch 1; headroom 4.6 GB.

Text input, FP16 KV cache, one card. Unified-memory capacity uses the GPU guide’s usable limit; negative headroom may require host offload. Speed source settings may differ from this 4K estimate.

Measured runs

No measured run for this pair. Sizing is an estimate, not a runtime test.

Quant memory at 4K

QuantTotal GBVerdictWeight sizing
GGUF Q4_075.4TightVerified file
GGUF Q8_0134.3Does not fitVerified file

Context capacity · GGUF Q4_0

ContextTotal GBVerdict
4K75.4Tight
16K75.5Tight
32K75.7Tight
128K76.5Tight

Generation speed

GGUF Q4_0: 284 tok/s · Bandwidth ceiling (not measured)

Other model / GPU pairs