Nemotron 3.5 30B-A3B

GPU model

Capacity range

Total VRAM

Practice

Model

Quantization

Context length

Reported speed

Runtime

Tools / workflow

Record origin

EXPLORE BY CAPACITY

Start with your machine.

Capacity is the reported total for this setup. Multi-GPU and unified-memory layouts can behave differently.

8 GB or lessLaptops, iGPUs and modest cards 9–12 GBEntry-level local inference 13–16 GBCurrent mid-range cards 17–20 GBRX 7900 XT and compact pro cards 21–24 GB3090, 4090 and 7900 XTX class 25–32 GB5090 and workstation cards 33–48 GBWorkstation and multi-GPU setups 49–96 GBLarge multi-GPU and datacenter setups 97 GB+Unified memory, servers and clusters

1 resultsEvery record keeps its origin visible.

Research & analysisGenesis archive2026-10

Dual Radeon Instinct MI50 benchmarks: dense models at 12 to 18 tok/s, MoE models at 47 to 61 tok/s

Two Radeon Instinct MI50 cards limited to 145W each are benchmarked with llama.cpp Vulkan across dense and MoE models, with MoE models far ahead of dense ones.

32 GB totalQwen 3.8 27B Gemma 4 31B llama.cpp AMD Instinct MI50

Cloud retained↑ 1

Setups running
Nemotron 3.5 30B-A3B.

What the catalog reports

Start with your machine.

Setups runningNemotron 3.5 30B-A3B.

What the catalog reports

Start with your machine.

Setups running
Nemotron 3.5 30B-A3B.