family llama · 7 variants · unknown
| benchmark | category | best score |
|---|---|---|
| MATH | math | 91.9 |
| variant | kind | quant | bits | vRAM | dl | org | year |
|---|---|---|---|---|---|---|---|
| Llama-3_3-Nemotron-Super-49B-v1 | original | bf16 | 16.0 | 181,588 | nvidia | 2025 | |
| nvidia_Llama-3_3-Nemotron-Super-49B-v1-GGUF | gguf | gguf | 2.0 | 1,668 | bartowski | 2025 | |
| Llama-3_3-Nemotron-Super-49B-v1-GGUF | gguf | gguf | 16.0 | 1,184 | unsloth | 2025 | |
| Llama-3_3-Nemotron-Super-49B-v1-FP8 | fp8 | fp8 | 8.0 | 873 | nvidia | 2025 | |
| Llama-3_3-Nemotron-Super-49B-GenRM | original | bf16 | 16.0 | 104 | nvidia | 2025 | |
| Llama-3_3-Nemotron-Super-49B-v1-GGUF | gguf | gguf | 16.0 | 64 | Mungert | 2025 | |
| Llama-3_3-Nemotron-Super-49B-GenRM-Multilingual | original | bf16 | 16.0 | 54 | nvidia | 2025 |