Model comparison
Llama 2-70B vs Nemotron 4 340b Instruct
Nemotron 4 340b Instruct is the stronger model overall, scoring 35.9 to 24.4 on the Noometry Index.
Last verified . 17 shared benchmarks.
Meta
NVIDIASummary
- They share 17 benchmarks with published results for both. Llama 2-70B scores higher in 0 categories and Nemotron 4 340b Instruct in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Nemotron 4 340b Instruct leads 34.3 to 8.1.
Side by side
| Llama 2-70B | Nemotron 4 340b Instruct | |
|---|---|---|
| Provider | Meta | NVIDIA |
| Noometry Index | 24.4 | 35.9 |
| Released | 2023-07-18 | 2024-06-14 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 35 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Nemotron 4 340b Instruct leads
Llama 2-70B: 31.4 (#286), Nemotron 4 340b Instruct: 35.2 (#228)
| Benchmark | Llama 2-70B | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Coding | 1079 | 1209 |
Reasoning Nemotron 4 340b Instruct leads
Llama 2-70B: 14.4 (#325), Nemotron 4 340b Instruct: 23.4 (#178)
| Benchmark | Llama 2-70B | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1202 |
| DTBench | 41.6% | — |
| BIG-Bench Hard | 64.9% | — |
| CommonsenseQA 2.0 | 50% | — |
| Epoch Capabilities Index | 113.79 | — |
| ForecastBench | 51.4 | — |
| HellaSwag | 85.3% | — |
| LAMBADA | 78.9% | — |
| PIQA | 82.8% | — |
| WinoGrande | 80.2% | — |
Math Nemotron 4 340b Instruct leads
Llama 2-70B: 8.1 (#326), Nemotron 4 340b Instruct: 34.3 (#195)
| Benchmark | Llama 2-70B | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Math | 1091 | 1216 |
| OTIS Mock AIME 2024-2025 | 0% | — |
| MATH Level 5 | 3.3% | — |
| GSM8K | 69.6% | — |
Knowledge Nemotron 4 340b Instruct leads
Llama 2-70B: 7.4 (#310), Nemotron 4 340b Instruct: 32.0 (#216)
| Benchmark | Llama 2-70B | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Expert | 1039 | 1171 |
| GPQA Diamond | 26.3% | — |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 88.6% | — |
| MMLU | 69.9% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |
Multilingual Nemotron 4 340b Instruct leads
Llama 2-70B: 27.7 (#274), Nemotron 4 340b Instruct: 37.9 (#233)
| Benchmark | Llama 2-70B | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Non-English | 1045 | 1206 |
| LMArena Chinese | 995 | 1214 |
| LMArena French | 1090 | 1197 |
| LMArena German | 1041 | 1192 |
| LMArena Japanese | 927 | 1128 |
| LMArena Korean | 964 | 1148 |
| LMArena Russian | 1083 | 1219 |
| LMArena Spanish | 1143 | 1195 |
Instruction Following Nemotron 4 340b Instruct leads
Llama 2-70B: 54.9 (#278), Nemotron 4 340b Instruct: 63.1 (#232)
| Benchmark | Llama 2-70B | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Instruction Following | 1071 | 1203 |
Long Context Nemotron 4 340b Instruct leads
Llama 2-70B: 32.3 (#270), Nemotron 4 340b Instruct: 37.1 (#221)
| Benchmark | Llama 2-70B | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Longer Query | 1062 | 1225 |
Writing & Preference Nemotron 4 340b Instruct leads
Llama 2-70B: 32.3 (#279), Nemotron 4 340b Instruct: 42.7 (#233)
| Benchmark | Llama 2-70B | Nemotron 4 340b Instruct |
|---|---|---|
| LMArena Text | 1115 | 1225 |
| LMArena Creative Writing | 1075 | 1203 |
| LMArena Multi-Turn | 1088 | 1214 |
Frequently asked questions
Is Llama 2-70B better than Nemotron 4 340b Instruct?
Nemotron 4 340b Instruct is the stronger model overall, scoring 35.9 to 24.4 on the Noometry Index.
Is Llama 2-70B or Nemotron 4 340b Instruct better for coding?
Nemotron 4 340b Instruct scores higher on coding benchmarks: 35.2 versus 31.4 in the Noometry coding category.
How many benchmarks do Llama 2-70B and Nemotron 4 340b Instruct share?
17 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and Nemotron 4 340b Instruct has 17.