Outmake Models
Models

Llama 3 3 Nemotron Super 49B V1

NVIDIA · March 2025

Get Hugging Face

A 50B chat model from NVIDIA.

Rating
3.3
Downloads
2.7K
Likes
52
Parameters
50B
Context
128K
Licence
other

Too big for this Mac

Its smallest size needs 29 GB of memory, more than M4 · 16 GB can give it.

Ratings

3.3 out of 5, across people's blind votes on its answers.

Text
1328

LM Arena ratings of October 28, 2026 (CC BY 4.0): people vote between two answers without knowing which model wrote which. The bar ends at 1488, the best any open model has. The Arena tests the full model; a small size on a Mac answers a little worse.

Llama 3 3 Nemotron Super 49B V1 runs well from a Mac with an M2 and 24 GB of memory. 47 of 59 Mac configurations run it well.

Compare with

The nearest models in size.

Sizes

From bartowski/nvidia_Llama-3_3-Nemotron-Super-49B-v1-GGUF on Hugging Face. A smaller size is a smaller download and a little less accurate.

SizeDownloadMemory at 8KGet
IQ2_XXS14 GB15 GBestimatedOpen
IQ2_XS15 GB16 GBestimatedOpen
IQ2_S16 GB17 GBestimatedOpen
IQ2_M17 GB18 GBestimatedOpen
Q2_K19 GB20 GBestimatedOpen
IQ3_XXS20 GB21 GBestimatedOpen
Q2_K_L20 GB21 GBestimatedOpen
IQ3_XS21 GB22 GBestimatedOpen
Q3_K_S22 GB23 GBestimatedOpen
IQ3_M23 GB24 GBestimatedOpen
Q3_K_M24 GB25 GBestimatedOpen
Q3_K_L26 GB27 GBestimatedOpen
IQ4_XS27 GB28 GBestimatedOpen
Q3_K_XL27 GB28 GBestimatedOpen
IQ4_NL28 GB29 GBestimatedOpen
Q4_028 GB30 GBestimatedOpen
Q4_K_S29 GB30 GBestimatedOpen
Q4_K_M30 GB31 GBestimatedOpen
Q4_K_L31 GB32 GBestimatedOpen
Q4_131 GB32 GBestimatedOpen
Q5_K_S34 GB35 GBestimatedOpen
Q5_K_M35 GB36 GBestimatedOpen
Q5_K_L36 GB37 GBestimatedOpen
Q6_K41 GB42 GBestimatedOpen
Q8_053 GB54 GBestimatedOpen
BF16100 GB101 GBestimatedOpen

Run It Anywhere

Llama 3 3 Nemotron Super 49B V1 is an open model, so any app that runs GGUF files can run it. These take the IQ2_XXS size from bartowski straight from Hugging Face.

  • LM StudioMac, Windows, LinuxOpen
  • JanMac, Windows, LinuxOpen
  • MstyMac, Windows, LinuxOpen
  • OllamaMac, Windows, Linuxollama run hf.co/bartowski/nvidia_Llama-3_3-Nemotron-Super-49B-v1-GGUF:IQ2_XXS
  • llama.cppMac, Windows, Linuxllama-server -hf bartowski/nvidia_Llama-3_3-Nemotron-Super-49B-v1-GGUF:IQ2_XXS
  • Docker Model RunnerMac, Windows, Linuxdocker model run hf.co/bartowski/nvidia_Llama-3_3-Nemotron-Super-49B-v1-GGUF:IQ2_XXS
On a PC Graphics Card
Video memoryLargest size that fitsDownloadMemory at 8K
6 GBNone worth runningneeds 21 GB
8 GBNone worth runningneeds 21 GB
10 GBNone worth runningneeds 21 GB
11 GBNone worth runningneeds 21 GB
12 GBNone worth runningneeds 21 GB
16 GBNone worth runningneeds 21 GB
20 GBNone worth runningneeds 21 GB
24 GBQ3_K_S22 GB23 GB
32 GBQ4_K_M30 GB31 GB
48 GBQ6_K41 GB42 GB

The largest size, of those that keep three bits a weight or more, whose memory at an 8K chat is under 92% of the card's; the rest is the desktop's. Estimated from the files, the same way as for a Mac; speed on a PC is not estimated yet. A size that doesn't fit the card still runs, slower, with part of it in ordinary memory.

On every Mac
MacSizeFitMemoryWrites
M6 · 16 GBIQ2_XXSToo big29 GB
M6 · 24 GBIQ2_XXSRuns well16 GB4.7
M6 · 32 GBQ2_K_LRuns well21 GB3.6
M5 · 16 GBIQ2_XXSToo big29 GB
M5 · 24 GBIQ2_XXSRuns well16 GB4.3
M5 · 32 GBQ2_K_LRuns well21 GB3.3
M5 Pro · 24 GBIQ2_XXSRuns well16 GB8.4
M5 Pro · 48 GBQ4_K_MRuns well31 GB4.7
M5 Pro · 64 GBQ4_K_SRuns great31 GB4.9
M5 Max · 36 GBIQ3_MRuns well24 GB8.7
M5 Max · 48 GBQ4_K_MRuns well31 GB9.1
M5 Max · 64 GBQ4_K_SRuns great31 GB9.5
M5 Max · 128 GBQ8_0Runs great68 GB6.8
M5 Ultra · 96 GBQ6_KRuns great45 GB14
M5 Ultra · 256 GBQ8_0Runs great68 GB13
M5 Ultra · 512 GBQ8_0Runs great68 GB13
M4 · 16 GBIQ2_XXSToo big29 GB
M4 · 24 GBIQ2_XXSRuns well16 GB3.3
M4 · 32 GBQ2_K_LRuns well21 GB2.6
M4 Pro · 24 GBIQ2_XXSRuns well16 GB7
M4 Pro · 48 GBQ4_K_MRuns well31 GB4.1
M4 Pro · 64 GBQ4_K_SRuns great31 GB4.2
M4 Max · 36 GBIQ3_MRuns well24 GB7.2
M4 Max · 48 GBQ4_K_MRuns well31 GB7.6
M4 Max · 64 GBQ4_K_SRuns great31 GB7.9
M4 Max · 96 GBQ6_KRuns great45 GB4.7
M4 Max · 128 GBQ8_0Runs great68 GB5.6
M3 · 8 GBIQ2_XXSToo big29 GB
M3 · 16 GBIQ2_XXSToo big29 GB
M3 · 24 GBIQ2_XXSRuns well16 GB2.8
M3 Pro · 18 GBIQ2_XXSToo big29 GB
M3 Pro · 36 GBIQ3_MRuns well24 GB2.9
M3 Max · 36 GBIQ3_MRuns well24 GB5.6
M3 Max · 48 GBQ4_K_MRuns well31 GB5.9
M3 Max · 64 GBQ4_K_SRuns great31 GB6.1
M3 Max · 96 GBQ6_KRuns great45 GB3.6
M3 Max · 128 GBQ8_0Runs great68 GB4.3
M3 Ultra · 96 GBQ6_KRuns great45 GB8.7
M3 Ultra · 256 GBQ8_0Runs great68 GB8.1
M3 Ultra · 512 GBQ8_0Runs great68 GB8.1
M2 · 8 GBIQ2_XXSToo big29 GB
M2 · 16 GBIQ2_XXSToo big29 GB
M2 · 24 GBIQ2_XXSRuns well16 GB2.8
M2 Pro · 16 GBIQ2_XXSToo big29 GB
M2 Pro · 32 GBQ2_K_LRuns well21 GB4.2
M2 Max · 32 GBQ2_K_LRuns well21 GB7.9
M2 Max · 64 GBQ4_K_SRuns great31 GB6.1
M2 Max · 96 GBQ6_KRuns great45 GB4.6
M2 Ultra · 64 GBQ4_K_SRuns great31 GB11
M2 Ultra · 128 GBQ8_0Runs great68 GB8.1
M2 Ultra · 192 GBQ8_0Runs great68 GB8.1
M1 · 8 GBIQ2_XXSToo big29 GB
M1 · 16 GBIQ2_XXSToo big29 GB
M1 Pro · 16 GBIQ2_XXSToo big29 GB
M1 Pro · 32 GBQ2_K_LRuns well21 GB4
M1 Max · 32 GBQ2_K_LRuns well21 GB7.3
M1 Max · 64 GBQ4_K_SRuns great31 GB5.7
M1 Ultra · 64 GBQ4_K_SRuns great31 GB9.9
M1 Ultra · 128 GBQ8_0Runs great68 GB7.5

Writes is words a second, estimated from the Mac's memory speed and the model's size; how, and how wrong.

Information

Maker
NVIDIA
Released
March 2025
Parameters
50B
Longest chat
128K tokens
Reads pictures
No
Licence
other
Downloads this month
2,720
Likes
52
Sizes
26, 14 to 100 GB
Kind
Maker's release

Badge for the model card

Shows on Llama 3 3 Nemotron Super 49B V1's Hugging Face card which Macs run it well, linked to this page. It updates every morning.

Runs on Mac badge

Files, downloads and likes read from Hugging Face on November 5, 2026 and refreshed every morning. Memory and speed are estimates until someone running Outmake measures this model on that Mac.

Your Computer

Every model is judged for this Mac. Both are under About This Mac, in the Apple menu.

Every Mac · Every graphics card

Civitai

Open in OutmakeHugging Face

New to Outmake? Download the free beta