Model comparison
DeepSeek-V3.1 vs ERNIE 5.0 0110
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 41.8 on the Noometry Index.
Last verified . 17 shared benchmarks.
DeepSeek
BaiduSummary
- They share 17 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 2 categories and ERNIE 5.0 0110 in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where DeepSeek-V3.1 leads 27.9 to 17.0.
- DeepSeek-V3.1 has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V3.1 | ERNIE 5.0 0110 | |
|---|---|---|
| Provider | DeepSeek | Baidu |
| Noometry Index | 42.8 | 41.8 |
| Released | 2025-08-21 | — |
| Weights | Open | Proprietary |
| Context window | 164K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $0.95 | — |
| Results tracked | 27 | 20 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding ERNIE 5.0 0110 leads
DeepSeek-V3.1: 40.3 (#144), ERNIE 5.0 0110: 43.0 (#94)
| Benchmark | DeepSeek-V3.1 | ERNIE 5.0 0110 |
|---|---|---|
| LMArena Coding | 1417 | 1455 |
| WeirdML | 38.4% | — |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.9 (#110), ERNIE 5.0 0110: 17.0 (#297)
| Benchmark | DeepSeek-V3.1 | ERNIE 5.0 0110 |
|---|---|---|
| LMArena Hard Prompts | 1417 | 1445 |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| NYT Connections (extended) | — | 10.3% |
| Thematic Generalization | — | 41.7% |
| DTBench | 82.7% | — |
| LMCA | 24.3% | — |
| Epoch Capabilities Index | 139.92 | — |
| ForecastBench | 58 | — |
Math Too close to call
DeepSeek-V3.1: 38.9 (#122), ERNIE 5.0 0110: 39.3 (#110)
| Benchmark | DeepSeek-V3.1 | ERNIE 5.0 0110 |
|---|---|---|
| LMArena Math | 1420 | 1437 |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.7 (#90), ERNIE 5.0 0110: 39.8 (#128)
| Benchmark | DeepSeek-V3.1 | ERNIE 5.0 0110 |
|---|---|---|
| LMArena Expert | 1405 | 1428 |
| Vectara Hallucination Rate | 5.5% | — |
Multimodal Not comparable
DeepSeek-V3.1: —, ERNIE 5.0 0110: 39.9 (#53)
| Benchmark | DeepSeek-V3.1 | ERNIE 5.0 0110 |
|---|---|---|
| LMArena Vision | — | 1249 |
Multilingual ERNIE 5.0 0110 leads
DeepSeek-V3.1: 51.6 (#106), ERNIE 5.0 0110: 54.1 (#49)
| Benchmark | DeepSeek-V3.1 | ERNIE 5.0 0110 |
|---|---|---|
| LMArena Non-English | 1400 | 1436 |
| LMArena Chinese | 1469 | 1512 |
| LMArena French | 1447 | 1467 |
| LMArena German | 1411 | 1460 |
| LMArena Japanese | 1378 | 1382 |
| LMArena Korean | 1337 | 1406 |
| LMArena Russian | 1405 | 1446 |
| LMArena Spanish | 1431 | 1473 |
Instruction Following Too close to call
DeepSeek-V3.1: 73.9 (#110), ERNIE 5.0 0110: 74.5 (#92)
| Benchmark | DeepSeek-V3.1 | ERNIE 5.0 0110 |
|---|---|---|
| LMArena Instruction Following | 1400 | 1413 |
Long Context ERNIE 5.0 0110 leads
DeepSeek-V3.1: 36.3 (#232), ERNIE 5.0 0110: 43.4 (#95)
| Benchmark | DeepSeek-V3.1 | ERNIE 5.0 0110 |
|---|---|---|
| LMArena Longer Query | 1422 | 1422 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference ERNIE 5.0 0110 leads
DeepSeek-V3.1: 60.3 (#98), ERNIE 5.0 0110: 63.1 (#66)
| Benchmark | DeepSeek-V3.1 | ERNIE 5.0 0110 |
|---|---|---|
| LMArena Text | 1420 | 1445 |
| LMArena Creative Writing | 1401 | 1426 |
| LMArena Multi-Turn | 1408 | 1434 |
| EQ-Bench Creative Writing | 1436 | — |
Frequently asked questions
Is DeepSeek-V3.1 better than ERNIE 5.0 0110?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 41.8 on the Noometry Index.
Is DeepSeek-V3.1 or ERNIE 5.0 0110 better for coding?
ERNIE 5.0 0110 scores higher on coding benchmarks: 43.0 versus 40.3 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1 and ERNIE 5.0 0110 share?
17 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and ERNIE 5.0 0110 has 20.