Memory at each precision
| Version | Weights | With context | Smallest Mac |
|---|---|---|---|
| 8-bit | 128 GB | 138 GB | 192 GB |
| 6-bit | 97.5 GB | 105 GB | 192 GB |
| 4-bit | 67.5 GB | 72.9 GB | 128 GB |
| 3-bit lower quality | 52.5 GB | 56.7 GB | 96 GB |
| 2-bit lower quality | 37.5 GB | 40.5 GB | 64 GB |
On every Apple chip
The smallest memory size of each chip that runs Nemotron 3.5 Super VL 120B A12B, the best version that fits it, and the writing speed you can expect.
| Chip | Needs | Version | Tokens/s |
|---|---|---|---|
| M6 | no size fits | Too big | – |
| M5 Ultra | 256 GB | 8-bit Runs well | 42–75 |
| M5 Max | 128 GB | 4-bit Runs well | 41–72 |
| M5 Pro | 64 GB | 2-bit Runs well | 36–65 |
| M5 | no size fits | Too big | – |
| M4 Max | 128 GB | 4-bit Runs well | 36–64 |
| M4 Pro | 64 GB | 2-bit Runs well | 32–57 |
| M4 | no size fits | Too big | – |
| M3 Ultra | 256 GB | 8-bit Runs well | 29–51 |
| M3 Max | 128 GB | 4-bit Runs well | 26–47 |
| M3 Pro | no size fits | Too big | – |
| M3 | no size fits | Too big | – |
| M2 Ultra | 128 GB | 4-bit Runs well | 53–94 |
| M2 Max | 64 GB | 2-bit Runs well | 47–84 |
| M2 Pro | no size fits | Too big | – |
| M2 | no size fits | Too big | – |
| M1 Ultra | 128 GB | 4-bit Runs well | 53–94 |
| M1 Max | 64 GB | 2-bit Runs well | 47–84 |
| M1 Pro | no size fits | Too big | – |
| M1 | no size fits | Too big | – |
How to run it
Step by step: download and run LLMs on a Mac · MLX vs Ollama vs llama.cpp
Prefer an app to the Terminal? Compare the Mac apps that run local models in the AI App Store: price, licence and where your data goes.
About Nemotron 3.5 Super VL 120B A12B
| Made by | NVIDIA |
|---|---|
| Parameters | 120B total · 12B active (mixture of experts) |
| Context | 1M tokens |
| Licence | OpenMDW-1.1 |
| Released | 2026-10 |
| Input | Text and images |
| Intelligence | 26.5 on the Artificial Analysis Intelligence Index |
| Weights | nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 |
Questions
- How much memory does Nemotron 3.5 Super VL 120B A12B need?
- About 67.5 GB of weights at 4-bit, plus room for the context: 72.9 GB in all. On a Mac that means 128 GB of unified memory or more, since macOS lets the GPU use about 75% of it by default.
- How fast is Nemotron 3.5 Super VL 120B A12B on a Mac?
- Writing speed depends on memory bandwidth. The fastest Mac for it here is an M2 Ultra with 128 GB at about 53–94 tokens a second (4-bit). The smallest Mac that runs it, with 128 GB, manages about 53–94 on an M2 Ultra.
- Why does Nemotron 3.5 Super VL 120B A12B run faster than its size suggests?
- It is a mixture-of-experts model: it holds 120B parameters in memory but reads only 12B for each token, so it writes at the speed of a much smaller model.
Similar models
For your model card or README
Shows the smallest Mac that runs Nemotron 3.5 Super VL 120B A12B well and links here.
[](https://mlx.app/run/nemotron-3.5-super-vl-120b-a12b)How it’s worked out
Memory: the model’s parameters at 8, 6, 4, 3 or 2 bits per weight plus MLX’s group scales (half a bit per weight), or its native format, with room for the context on top. By default macOS lets the GPU use about 75% of unified memory. Speed: memory bandwidth divided by the bytes read per token is the ceiling for one conversation; in llama.cpp’s public Apple Silicon benchmarks real runs reach about 45% to 80% of it, which is the range shown. These are estimates, not measurements: measure your own Mac on MLX Bench. Model figures from each official model card, chip figures from Apple’s Tech Specs. Checked 2026-10-19. Same data as mlx.app/compare.
Open data: mlx.app/data (JSON, CSV)