Model comparison
ERNIE 5.0 0110 vs Llama 2-7B
ERNIE 5.0 0110 is the stronger model overall, scoring 41.8 to 29.1 on the Noometry Index.
Last verified . 15 shared benchmarks.
Baidu
MetaSummary
- They share 15 benchmarks with published results for both. ERNIE 5.0 0110 scores higher in 8 categories and Llama 2-7B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where ERNIE 5.0 0110 leads 63.1 to 28.0.
- Llama 2-7B has downloadable open weights; the other is API-only.
Side by side
| ERNIE 5.0 0110 | Llama 2-7B | |
|---|---|---|
| Provider | Baidu | Meta |
| Noometry Index | 41.8 | 29.1 |
| Released | — | 2023-07-18 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 20 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding ERNIE 5.0 0110 leads
ERNIE 5.0 0110: 43.0 (#94), Llama 2-7B: 29.2 (#307)
| Benchmark | ERNIE 5.0 0110 | Llama 2-7B |
|---|---|---|
| LMArena Coding | 1455 | 1002 |
Reasoning ERNIE 5.0 0110 leads
ERNIE 5.0 0110: 17.0 (#297), Llama 2-7B: 15.7 (#312)
| Benchmark | ERNIE 5.0 0110 | Llama 2-7B |
|---|---|---|
| LMArena Hard Prompts | 1445 | 1009 |
| NYT Connections (extended) | 10.3% | — |
| Chess Puzzles | — | 0% |
| Thematic Generalization | 41.7% | — |
| BIG-Bench Hard | — | 39.2% |
| Epoch Capabilities Index | — | 99.06 |
| HellaSwag | — | 77.2% |
| LAMBADA | — | 73.3% |
| PIQA | — | 78.8% |
| WinoGrande | — | 69.2% |
Math ERNIE 5.0 0110 leads
ERNIE 5.0 0110: 39.3 (#110), Llama 2-7B: 30.7 (#233)
| Benchmark | ERNIE 5.0 0110 | Llama 2-7B |
|---|---|---|
| LMArena Math | 1437 | 1042 |
| GSM8K | — | 16.7% |
Knowledge ERNIE 5.0 0110 leads
ERNIE 5.0 0110: 39.8 (#128), Llama 2-7B: 28.2 (#248)
| Benchmark | ERNIE 5.0 0110 | Llama 2-7B |
|---|---|---|
| LMArena Expert | 1428 | 1036 |
| ARC (AI2) Challenge | — | 45.9% |
| BoolQ | — | 77.9% |
| MMLU | — | 45.8% |
| OpenBookQA | — | 58.6% |
| TriviaQA | — | 73.7% |
Multimodal Not comparable
ERNIE 5.0 0110: 39.9 (#53), Llama 2-7B: —
| Benchmark | ERNIE 5.0 0110 | Llama 2-7B |
|---|---|---|
| LMArena Vision | 1249 | — |
| ScienceQA | — | 43.1% |
Multilingual ERNIE 5.0 0110 leads
ERNIE 5.0 0110: 54.1 (#49), Llama 2-7B: 23.8 (#293)
| Benchmark | ERNIE 5.0 0110 | Llama 2-7B |
|---|---|---|
| LMArena Non-English | 1436 | 973 |
| LMArena Chinese | 1512 | 973 |
| LMArena French | 1467 | 970 |
| LMArena German | 1460 | 978 |
| LMArena Russian | 1446 | 995 |
| LMArena Spanish | 1473 | 1007 |
| LMArena Japanese | 1382 | — |
| LMArena Korean | 1406 | — |
Instruction Following ERNIE 5.0 0110 leads
ERNIE 5.0 0110: 74.5 (#92), Llama 2-7B: 50.8 (#298)
| Benchmark | ERNIE 5.0 0110 | Llama 2-7B |
|---|---|---|
| LMArena Instruction Following | 1413 | 1006 |
Long Context ERNIE 5.0 0110 leads
ERNIE 5.0 0110: 43.4 (#95), Llama 2-7B: 30.4 (#287)
| Benchmark | ERNIE 5.0 0110 | Llama 2-7B |
|---|---|---|
| LMArena Longer Query | 1422 | 999 |
Writing & Preference ERNIE 5.0 0110 leads
ERNIE 5.0 0110: 63.1 (#66), Llama 2-7B: 28.0 (#298)
| Benchmark | ERNIE 5.0 0110 | Llama 2-7B |
|---|---|---|
| LMArena Text | 1445 | 1053 |
| LMArena Creative Writing | 1426 | 1033 |
| LMArena Multi-Turn | 1434 | 1029 |
Frequently asked questions
Is ERNIE 5.0 0110 better than Llama 2-7B?
ERNIE 5.0 0110 is the stronger model overall, scoring 41.8 to 29.1 on the Noometry Index.
Is ERNIE 5.0 0110 or Llama 2-7B better for coding?
ERNIE 5.0 0110 scores higher on coding benchmarks: 43.0 versus 29.2 in the Noometry coding category.
How many benchmarks do ERNIE 5.0 0110 and Llama 2-7B share?
15 benchmarks have published results for both models. ERNIE 5.0 0110 has 20 scored results on Noometry and Llama 2-7B has 29.