Model comparison
DBRX vs ERNIE 5.1
ERNIE 5.1 is the stronger model overall, scoring 43.8 to 29.4 on the Noometry Index.
Last verified . 17 shared benchmarks.
Databricks
BaiduSummary
- They share 17 benchmarks with published results for both. DBRX scores higher in 0 categories and ERNIE 5.1 in 8 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where ERNIE 5.1 leads 65.1 to 33.6.
- DBRX has downloadable open weights; the other is API-only.
Side by side
| DBRX | ERNIE 5.1 | |
|---|---|---|
| Provider | Databricks | Baidu |
| Noometry Index | 29.4 | 43.8 |
| Released | 2024-03-27 | — |
| Weights | Open | Proprietary |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 21 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding ERNIE 5.1 leads
DBRX: 32.9 (#266), ERNIE 5.1: 44.0 (#76)
| Benchmark | DBRX | ERNIE 5.1 |
|---|---|---|
| LMArena Coding | 1132 | 1488 |
| HumanEval+ | 70.1% | — |
| MBPP+ | 55.8% | — |
Agentic & Tool Use Not comparable
DBRX: —, ERNIE 5.1: —
| Benchmark | DBRX | ERNIE 5.1 |
|---|---|---|
| LMArena Search | — | 1227 |
Reasoning Too close to call
DBRX: 21.4 (#222), ERNIE 5.1: 21.9 (#211)
| Benchmark | DBRX | ERNIE 5.1 |
|---|---|---|
| LMArena Hard Prompts | 1113 | 1481 |
| NYT Connections (extended) | — | 23.4% |
Math ERNIE 5.1 leads
DBRX: 24.3 (#269), ERNIE 5.1: 40.3 (#92)
| Benchmark | DBRX | ERNIE 5.1 |
|---|---|---|
| LMArena Math | 1145 | 1481 |
| MATH Level 5 | 11.7% | — |
Knowledge ERNIE 5.1 leads
DBRX: 14.9 (#294), ERNIE 5.1: 41.9 (#102)
| Benchmark | DBRX | ERNIE 5.1 |
|---|---|---|
| LMArena Expert | 1076 | 1492 |
| GPQA Diamond | 32.9% | — |
Multilingual ERNIE 5.1 leads
DBRX: 29.3 (#268), ERNIE 5.1: 55.5 (#29)
| Benchmark | DBRX | ERNIE 5.1 |
|---|---|---|
| LMArena Non-English | 1071 | 1454 |
| LMArena Chinese | 1068 | 1508 |
| LMArena French | 1096 | 1488 |
| LMArena German | 1057 | 1470 |
| LMArena Japanese | 990 | 1422 |
| LMArena Korean | 993 | 1427 |
| LMArena Russian | 1078 | 1459 |
| LMArena Spanish | 1064 | 1473 |
Instruction Following ERNIE 5.1 leads
DBRX: 57.5 (#270), ERNIE 5.1: 76.7 (#37)
| Benchmark | DBRX | ERNIE 5.1 |
|---|---|---|
| LMArena Instruction Following | 1112 | 1460 |
Long Context ERNIE 5.1 leads
DBRX: 33.7 (#258), ERNIE 5.1: 44.7 (#59)
| Benchmark | DBRX | ERNIE 5.1 |
|---|---|---|
| LMArena Longer Query | 1112 | 1462 |
Writing & Preference ERNIE 5.1 leads
DBRX: 33.6 (#275), ERNIE 5.1: 65.1 (#52)
| Benchmark | DBRX | ERNIE 5.1 |
|---|---|---|
| LMArena Text | 1119 | 1468 |
| LMArena Creative Writing | 1104 | 1441 |
| LMArena Multi-Turn | 1111 | 1471 |
Frequently asked questions
Is DBRX better than ERNIE 5.1?
ERNIE 5.1 is the stronger model overall, scoring 43.8 to 29.4 on the Noometry Index.
Is DBRX or ERNIE 5.1 better for coding?
ERNIE 5.1 scores higher on coding benchmarks: 44.0 versus 32.9 in the Noometry coding category.
How many benchmarks do DBRX and ERNIE 5.1 share?
17 benchmarks have published results for both models. DBRX has 21 scored results on Noometry and ERNIE 5.1 has 19.