← CanIRunLocal · All hardware · All models
No. Nemotron 3.5 Super 120B-A12B (MoE) does not fit on RTX 4090 24GB: Q4_K_M needs 92.7 GB and 23.5 GB is usable.
Hardware: RTX 4090 24GB · Model: Nemotron 3.5 Super 120B-A12B (MoE)
RTX 4090 24GB has 24 GB of VRAM and 1008 GB/s of memory bandwidth. The checker holds back 0.5 GB of VRAM for the driver and assumes 16 GB of system RAM, so fast memory usable by Ollama is 23.5 GB. Partial offload may also use 60% of that system RAM. Nemotron 3.5 Super 120B-A12B (MoE) (120B, 12B active, NVIDIA) at Q4_K_M is a 90 GB download and needs 92.7 GB with headroom for about 8k tokens.
| Best quant | Q4_K_M |
|---|---|
| Download size | 90 GB |
| Memory needed | 92.7 GB |
| Usable fast memory | 23.5 GB |
| Fit | does not fit |
| Estimated speed | no estimate, it does not fit |
| Command | ollama run nemotron-3.5-super:120b-a12b-q4_K_M |
Smallest hardware that runs it: M2 Ultra 128GB at Q4_K_M (fits in fast memory, 92.7 GB needed, 96.0 GB usable, about 36 tok/s (estimate)). Can a M2 Ultra 128GB run Nemotron 3.5 Super 120B-A12B (MoE)?
A smaller NVIDIA model that fits this hardware: Nemotron 3.5 Bolt 74B-A9B (MoE, IQ2-class) at Default (partial offload (slower), 27.7 GB). Can a RTX 4090 24GB run Nemotron 3.5 Bolt 74B-A9B (MoE, IQ2-class)?
Open RTX 4090 24GB in the checkerEstimates, not benchmarks. Model sizes are ollama.com download sizes; memory needed adds headroom for ~8k tokens of context; speeds are rough figures from memory bandwidth. Data updated 2026-10-23.
A project by OnDevice Labs · llms.txt