Model comparison
Deepseek Coder v2 vs ERNIE 5.1
ERNIE 5.1 is the stronger model overall, scoring 43.8 to 35.9 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeek
BaiduSummary
- They share 17 benchmarks with published results for both. Deepseek Coder v2 scores higher in 1 category and ERNIE 5.1 in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where ERNIE 5.1 leads 65.1 to 38.2.
- Deepseek Coder v2 has downloadable open weights; the other is API-only.
Side by side
| Deepseek Coder v2 | ERNIE 5.1 | |
|---|---|---|
| Provider | DeepSeek | Baidu |
| Noometry Index | 35.9 | 43.8 |
| Released | 2024-06-17 | — |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 24 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding ERNIE 5.1 leads
Deepseek Coder v2: 38.1 (#183), ERNIE 5.1: 44.0 (#76)
| Benchmark | Deepseek Coder v2 | ERNIE 5.1 |
|---|---|---|
| LMArena Coding | 1251 | 1488 |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 59.7% | — |
| HumanEval+ | 82.3% | — |
| MBPP+ | 75.1% | — |
Agentic & Tool Use Not comparable
Deepseek Coder v2: —, ERNIE 5.1: —
| Benchmark | Deepseek Coder v2 | ERNIE 5.1 |
|---|---|---|
| LMArena Search | — | 1227 |
Reasoning Deepseek Coder v2 leads
Deepseek Coder v2: 23.6 (#176), ERNIE 5.1: 21.9 (#211)
| Benchmark | Deepseek Coder v2 | ERNIE 5.1 |
|---|---|---|
| LMArena Hard Prompts | 1207 | 1481 |
| NYT Connections (extended) | — | 23.4% |
| WinoGrande | 83.7% | — |
Math ERNIE 5.1 leads
Deepseek Coder v2: 34.9 (#190), ERNIE 5.1: 40.3 (#92)
| Benchmark | Deepseek Coder v2 | ERNIE 5.1 |
|---|---|---|
| LMArena Math | 1241 | 1481 |
| GSM8K | 94.5% | — |
Knowledge ERNIE 5.1 leads
Deepseek Coder v2: 32.3 (#212), ERNIE 5.1: 41.9 (#102)
| Benchmark | Deepseek Coder v2 | ERNIE 5.1 |
|---|---|---|
| LMArena Expert | 1181 | 1492 |
| ARC (AI2) Challenge | 64.3% | — |
Multilingual ERNIE 5.1 leads
Deepseek Coder v2: 36.3 (#240), ERNIE 5.1: 55.5 (#29)
| Benchmark | Deepseek Coder v2 | ERNIE 5.1 |
|---|---|---|
| LMArena Non-English | 1182 | 1454 |
| LMArena Chinese | 1201 | 1508 |
| LMArena French | 1185 | 1488 |
| LMArena German | 1164 | 1470 |
| LMArena Japanese | 1126 | 1422 |
| LMArena Korean | 1104 | 1427 |
| LMArena Russian | 1188 | 1459 |
| LMArena Spanish | 1153 | 1473 |
Instruction Following ERNIE 5.1 leads
Deepseek Coder v2: 61.7 (#242), ERNIE 5.1: 76.7 (#37)
| Benchmark | Deepseek Coder v2 | ERNIE 5.1 |
|---|---|---|
| LMArena Instruction Following | 1180 | 1460 |
Long Context ERNIE 5.1 leads
Deepseek Coder v2: 37.0 (#224), ERNIE 5.1: 44.7 (#59)
| Benchmark | Deepseek Coder v2 | ERNIE 5.1 |
|---|---|---|
| LMArena Longer Query | 1219 | 1462 |
Writing & Preference ERNIE 5.1 leads
Deepseek Coder v2: 38.2 (#253), ERNIE 5.1: 65.1 (#52)
| Benchmark | Deepseek Coder v2 | ERNIE 5.1 |
|---|---|---|
| LMArena Text | 1191 | 1468 |
| LMArena Creative Writing | 1120 | 1441 |
| LMArena Multi-Turn | 1177 | 1471 |
Frequently asked questions
Is Deepseek Coder v2 better than ERNIE 5.1?
ERNIE 5.1 is the stronger model overall, scoring 43.8 to 35.9 on the Noometry Index.
Is Deepseek Coder v2 or ERNIE 5.1 better for coding?
ERNIE 5.1 scores higher on coding benchmarks: 44.0 versus 38.1 in the Noometry coding category.
How many benchmarks do Deepseek Coder v2 and ERNIE 5.1 share?
17 benchmarks have published results for both models. Deepseek Coder v2 has 24 scored results on Noometry and ERNIE 5.1 has 19.