Model comparison
Llama 2-70B vs Nemotron 3 Ultra
Nemotron 3 Ultra is the stronger model overall, scoring 44.7 to 24.1 on the Noometry Index.
Last verified . 20 shared benchmarks.
Summary
- They share 20 benchmarks with published results for both. Llama 2-70B scores higher in 0 categories and Nemotron 3 Ultra in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Nemotron 3 Ultra leads 52.5 to 7.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0% for Llama 2-70B and 86.7% for Nemotron 3 Ultra.
Side by side
| Llama 2-70B | Nemotron 3 Ultra | |
|---|---|---|
| Provider | Meta | NVIDIA |
| Noometry Index | 24.1 | 44.7 |
| Released | 2023-07-18 | 2026-06-04 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.60 |
| Output $ / M tokens | — | $2.40 |
| Results tracked | 35 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Nemotron 3 Ultra leads
Llama 2-70B: 30.8 (#281), Nemotron 3 Ultra: 37.7 (#174)
| Benchmark | Llama 2-70B | Nemotron 3 Ultra |
|---|---|---|
| LMArena Coding | 1079 | 1468 |
| FrontierCode | — | 13.6% |
| SciCode | — | 40.3% |
| WeirdML | — | 43.5% |
Reasoning Nemotron 3 Ultra leads
Llama 2-70B: 13.7 (#324), Nemotron 3 Ultra: 30.3 (#78)
| Benchmark | Llama 2-70B | Nemotron 3 Ultra |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1452 |
| DTBench | 41.6% | 90.1% |
| Epoch Capabilities Index | 113.59 | 146.2 |
| CritPt | — | 3.1% |
| Chess Puzzles | — | 12% |
| Mystery Game Puzzles | — | 20% |
| LMCA | — | 36.9% |
| BIG-Bench Hard | 64.9% | — |
| CommonsenseQA 2.0 | 50% | — |
| ForecastBench | 51.4 | — |
| HellaSwag | 85.3% | — |
| LAMBADA | 78.9% | — |
| PIQA | 82.8% | — |
| WinoGrande | 80.2% | — |
Math Nemotron 3 Ultra leads
Llama 2-70B: 7.5 (#320), Nemotron 3 Ultra: 34.4 (#160)
| Benchmark | Llama 2-70B | Nemotron 3 Ultra |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0% | 86.7% |
| LMArena Math | 1091 | 1456 |
| ProofBench | — | 2% |
| MATH Level 5 | 3.3% | — |
| GSM8K | 69.6% | — |
Knowledge Nemotron 3 Ultra leads
Llama 2-70B: 7.4 (#305), Nemotron 3 Ultra: 52.5 (#56)
| Benchmark | Llama 2-70B | Nemotron 3 Ultra |
|---|---|---|
| GPQA Diamond | 26.3% | 85.4% |
| LMArena Expert | 1039 | 1472 |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 88.6% | — |
| MMLU | 69.9% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |
Multilingual Nemotron 3 Ultra leads
Llama 2-70B: 27.9 (#269), Nemotron 3 Ultra: 53.7 (#61)
| Benchmark | Llama 2-70B | Nemotron 3 Ultra |
|---|---|---|
| LMArena Non-English | 1045 | 1427 |
| LMArena Chinese | 994 | 1498 |
| LMArena French | 1089 | 1475 |
| LMArena German | 1041 | 1471 |
| LMArena Korean | 964 | 1387 |
| LMArena Russian | 1083 | 1415 |
| LMArena Spanish | 1142 | 1452 |
| LMArena Japanese | 927 | — |
Instruction Following Nemotron 3 Ultra leads
Llama 2-70B: 55.1 (#273), Nemotron 3 Ultra: 74.9 (#84)
| Benchmark | Llama 2-70B | Nemotron 3 Ultra |
|---|---|---|
| LMArena Instruction Following | 1071 | 1417 |
Long Context Nemotron 3 Ultra leads
Llama 2-70B: 32.2 (#265), Nemotron 3 Ultra: 43.9 (#77)
| Benchmark | Llama 2-70B | Nemotron 3 Ultra |
|---|---|---|
| LMArena Longer Query | 1062 | 1437 |
Writing & Preference Nemotron 3 Ultra leads
Llama 2-70B: 32.5 (#274), Nemotron 3 Ultra: 66.3 (#35)
| Benchmark | Llama 2-70B | Nemotron 3 Ultra |
|---|---|---|
| LMArena Text | 1115 | 1444 |
| LMArena Creative Writing | 1075 | 1392 |
| LMArena Multi-Turn | 1088 | 1412 |
| EQ-Bench Creative Writing | — | 1692 |
Frequently asked questions
Is Llama 2-70B better than Nemotron 3 Ultra?
Nemotron 3 Ultra is the stronger model overall, scoring 44.7 to 24.1 on the Noometry Index.
Is Llama 2-70B or Nemotron 3 Ultra better for coding?
Nemotron 3 Ultra scores higher on coding benchmarks: 37.7 versus 30.8 in the Noometry coding category.
How many benchmarks do Llama 2-70B and Nemotron 3 Ultra share?
20 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and Nemotron 3 Ultra has 29.