unsloth / nemotron-3-super family / text-generation
nemotron-3-super-120b-a12b
nvidia/nemotron-3-super-120b-a12b
Run locally Smallest indexed GGUF / needs Ollama
ollama run hf.co/nvidia/nemotron-3-super-120b-a12b/UD-IQ1_M/NVIDIA-Nemotron-3-Super-120B-A12B-UD-IQ1_M-00001-of-00003.gguf Benchmarks
Where it scores.
General Provisional
68.8
Coding Provisional
53.7
Math Provisional
60.5
Reasoning Provisional
51.7
Normalized 0–100 across indexed models.
Runnable variants
Pick a size.
| Precision | Format | Size | Params | File |
|---|---|---|---|---|
| IQ1_M / smallest | GGUF | 52.7 GB | — | UD-IQ1_M/NVIDIA-Nemotron-3-Super-120B-A12B-UD-IQ1_M-00001-of-00003.gguf |
| IQ2_XXS | GGUF | 52.7 GB | — | UD-IQ2_XXS/NVIDIA-Nemotron-3-Super-120B-A12B-UD-IQ2_XXS-00001-of-00003.gguf |
| IQ2_M | GGUF | 52.7 GB | — | UD-IQ2_M/NVIDIA-Nemotron-3-Super-120B-A12B-UD-IQ2_M-00001-of-00003.gguf |
| Q2_K_XL | GGUF | 54.7 GB | — | UD-Q2_K_XL/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q2_K_XL-00001-of-00003.gguf |
| IQ3_S | GGUF | 56.6 GB | — | UD-IQ3_S/NVIDIA-Nemotron-3-Super-120B-A12B-UD-IQ3_S-00001-of-00003.gguf |
| IQ3_XXS | GGUF | 56.6 GB | — | UD-IQ3_XXS/NVIDIA-Nemotron-3-Super-120B-A12B-UD-IQ3_XXS-00001-of-00003.gguf |
| Q3_K_M | GGUF | 61.7 GB | — | UD-Q3_K_M/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q3_K_M-00001-of-00003.gguf |
| Q3_K_S | GGUF | 61.7 GB | — | UD-Q3_K_S/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q3_K_S-00001-of-00003.gguf |
| Q3_K_XL | GGUF | 62.6 GB | — | UD-Q3_K_XL/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q3_K_XL-00001-of-00003.gguf |
| IQ4_NL | GGUF | 64.5 GB | — | UD-IQ4_NL/NVIDIA-Nemotron-3-Super-120B-A12B-UD-IQ4_NL-00001-of-00003.gguf |
| IQ4_XS | GGUF | 64.5 GB | — | UD-IQ4_XS/NVIDIA-Nemotron-3-Super-120B-A12B-UD-IQ4_XS-00001-of-00003.gguf |
| Q4_K_S | GGUF | 79.0 GB | — | UD-Q4_K_S/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q4_K_S-00001-of-00003.gguf |
| Q4 | GGUF | 82.1 GB | — | MXFP4_MOE/NVIDIA-Nemotron-3-Super-120B-A12B-MXFP4_MOE-00001-of-00003.gguf |
| Q4_K_M | GGUF | 82.5 GB | — | UD-Q4_K_M/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q4_K_M-00001-of-00003.gguf |
| Q4_K_XL | GGUF | 83.8 GB | — | UD-Q4_K_XL/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q4_K_XL-00001-of-00003.gguf |
| Q5_K_S | GGUF | 89.7 GB | — | UD-Q5_K_S/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q5_K_S-00001-of-00003.gguf |
| Q5_K_M | GGUF | 107.3 GB | — | UD-Q5_K_M/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q5_K_M-00001-of-00004.gguf |
| Q5_K_XL | GGUF | 107.6 GB | — | UD-Q5_K_XL/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q5_K_XL-00001-of-00004.gguf |
| Q6_K | GGUF | 114.8 GB | — | UD-Q6_K/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q6_K-00001-of-00004.gguf |
| Q6_K_XL | GGUF | 117.9 GB | — | UD-Q6_K_XL/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q6_K_XL-00001-of-00004.gguf |
| Q8_0 | GGUF | 128.5 GB | — | Q8_0/NVIDIA-Nemotron-3-Super-120B-A12B-Q8_0-00001-of-00004.gguf |
| Q8_K_XL | GGUF | 132.5 GB | — | UD-Q8_K_XL/NVIDIA-Nemotron-3-Super-120B-A12B-UD-Q8_K_XL-00001-of-00004.gguf |
| BF16 | GGUF | 241.5 GB | — | BF16/NVIDIA-Nemotron-3-Super-120B-A12B-BF16-00001-of-00005.gguf |
Sizes are file sizes, not peak runtime memory.
Raw benchmarks
The inputs.
| Source | Task | Raw | Normalized |
|---|---|---|---|
| ZeroEval | AIME2025 | 90.2 | 90.2 |
| ZeroEval | GPQA | 82.7 | 82.7 |
| ZeroEval | HLE | 22.8 | 22.8 |
| ZeroEval | SWEBenchVerified | 53.7 | 53.7 |
| ZeroEval | SciCode | 42.0 | 42.0 |
From the Hugging Face model card
About this model.
From the Hugging Face model card