Model comparison
ERNIE 5.1 vs Llama 2-70B
ERNIE 5.1 is the stronger model overall, scoring 43.1 to 24.1 on the Noometry Index.
Last verified . 17 shared benchmarks.
Baidu
MetaSummary
- They share 17 benchmarks with published results for both. ERNIE 5.1 scores higher in 8 categories and Llama 2-70B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where ERNIE 5.1 leads 41.2 to 7.4.
- Llama 2-70B has downloadable open weights; the other is API-only.
Side by side
| ERNIE 5.1 | Llama 2-70B | |
|---|---|---|
| Provider | Baidu | Meta |
| Noometry Index | 43.1 | 24.1 |
| Released | — | 2023-07-18 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 19 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding ERNIE 5.1 leads
ERNIE 5.1: 43.4 (#78), Llama 2-70B: 30.8 (#281)
| Benchmark | ERNIE 5.1 | Llama 2-70B |
|---|---|---|
| LMArena Coding | 1488 | 1079 |
Agentic & Tool Use Not comparable
ERNIE 5.1: —, Llama 2-70B: —
| Benchmark | ERNIE 5.1 | Llama 2-70B |
|---|---|---|
| LMArena Search | 1227 | — |
Reasoning ERNIE 5.1 leads
ERNIE 5.1: 20.9 (#209), Llama 2-70B: 13.7 (#324)
| Benchmark | ERNIE 5.1 | Llama 2-70B |
|---|---|---|
| LMArena Hard Prompts | 1481 | 1073 |
| NYT Connections (extended) | 23.4% | — |
| DTBench | — | 41.6% |
| BIG-Bench Hard | — | 64.9% |
| CommonsenseQA 2.0 | — | 50% |
| Epoch Capabilities Index | — | 113.59 |
| ForecastBench | — | 51.4 |
| HellaSwag | — | 85.3% |
| LAMBADA | — | 78.9% |
| PIQA | — | 82.8% |
| WinoGrande | — | 80.2% |
Math ERNIE 5.1 leads
ERNIE 5.1: 37.5 (#96), Llama 2-70B: 7.5 (#320)
| Benchmark | ERNIE 5.1 | Llama 2-70B |
|---|---|---|
| LMArena Math | 1481 | 1091 |
| OTIS Mock AIME 2024-2025 | — | 0% |
| MATH Level 5 | — | 3.3% |
| GSM8K | — | 69.6% |
Knowledge ERNIE 5.1 leads
ERNIE 5.1: 41.2 (#103), Llama 2-70B: 7.4 (#305)
| Benchmark | ERNIE 5.1 | Llama 2-70B |
|---|---|---|
| LMArena Expert | 1491 | 1039 |
| GPQA Diamond | — | 26.3% |
| ARC (AI2) Challenge | — | 78.3% |
| BoolQ | — | 88.6% |
| MMLU | — | 69.9% |
| OpenBookQA | — | 60.2% |
| TriviaQA | — | 87.6% |
Multilingual ERNIE 5.1 leads
ERNIE 5.1: 55.7 (#28), Llama 2-70B: 27.9 (#269)
| Benchmark | ERNIE 5.1 | Llama 2-70B |
|---|---|---|
| LMArena Non-English | 1455 | 1045 |
| LMArena Chinese | 1507 | 994 |
| LMArena French | 1486 | 1089 |
| LMArena German | 1470 | 1041 |
| LMArena Japanese | 1422 | 927 |
| LMArena Korean | 1428 | 964 |
| LMArena Russian | 1459 | 1083 |
| LMArena Spanish | 1472 | 1142 |
Instruction Following ERNIE 5.1 leads
ERNIE 5.1: 76.9 (#35), Llama 2-70B: 55.1 (#273)
| Benchmark | ERNIE 5.1 | Llama 2-70B |
|---|---|---|
| LMArena Instruction Following | 1460 | 1071 |
Long Context ERNIE 5.1 leads
ERNIE 5.1: 44.7 (#56), Llama 2-70B: 32.2 (#265)
| Benchmark | ERNIE 5.1 | Llama 2-70B |
|---|---|---|
| LMArena Longer Query | 1462 | 1062 |
Writing & Preference ERNIE 5.1 leads
ERNIE 5.1: 65.3 (#50), Llama 2-70B: 32.5 (#274)
| Benchmark | ERNIE 5.1 | Llama 2-70B |
|---|---|---|
| LMArena Text | 1468 | 1115 |
| LMArena Creative Writing | 1440 | 1075 |
| LMArena Multi-Turn | 1471 | 1088 |
Frequently asked questions
Is ERNIE 5.1 better than Llama 2-70B?
ERNIE 5.1 is the stronger model overall, scoring 43.1 to 24.1 on the Noometry Index.
Is ERNIE 5.1 or Llama 2-70B better for coding?
ERNIE 5.1 scores higher on coding benchmarks: 43.4 versus 30.8 in the Noometry coding category.
How many benchmarks do ERNIE 5.1 and Llama 2-70B share?
17 benchmarks have published results for both models. ERNIE 5.1 has 19 scored results on Noometry and Llama 2-70B has 35.