Model comparison
ERNIE 5.1 vs Phi-4
ERNIE 5.1 is the stronger model overall, scoring 43.8 to 31.2 on the Noometry Index.
Last verified . 17 shared benchmarks.
Baidu
MicrosoftSummary
- They share 17 benchmarks with published results for both. ERNIE 5.1 scores higher in 8 categories and Phi-4 in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where ERNIE 5.1 leads 65.1 to 40.5.
- Phi-4 has downloadable open weights; the other is API-only.
Side by side
| ERNIE 5.1 | Phi-4 | |
|---|---|---|
| Provider | Baidu | Microsoft |
| Noometry Index | 43.8 | 31.2 |
| Released | — | 2024-12-11 |
| Weights | Proprietary | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.07 |
| Output $ / M tokens | — | $0.14 |
| Results tracked | 19 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding ERNIE 5.1 leads
ERNIE 5.1: 44.0 (#76), Phi-4: 34.4 (#239)
| Benchmark | ERNIE 5.1 | Phi-4 |
|---|---|---|
| LMArena Coding | 1488 | 1231 |
| BigCodeBench Instruct | — | 45.5% |
| LiveBench Coding | — | 30.7% |
| BigCodeBench Complete | — | 55.4% |
Agentic & Tool Use Not comparable
ERNIE 5.1: —, Phi-4: 22.8 (#128)
| Benchmark | ERNIE 5.1 | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |
| LMArena Search | 1227 | — |
Reasoning ERNIE 5.1 leads
ERNIE 5.1: 21.9 (#211), Phi-4: 17.7 (#291)
| Benchmark | ERNIE 5.1 | Phi-4 |
|---|---|---|
| LMArena Hard Prompts | 1481 | 1220 |
| NYT Connections (extended) | 23.4% | — |
| Chess Puzzles | — | 1% |
| LiveBench Reasoning | — | 47.8% |
| LiveBench Data Analysis | — | 45.2% |
| Epoch Capabilities Index | — | 130.42 |
| LiveBench | — | 41.6% |
Math ERNIE 5.1 leads
ERNIE 5.1: 40.3 (#92), Phi-4: 20.8 (#285)
| Benchmark | ERNIE 5.1 | Phi-4 |
|---|---|---|
| LMArena Math | 1481 | 1246 |
| OTIS Mock AIME 2024-2025 | — | 13.8% |
| LiveBench Math | — | 42% |
| MATH Level 5 | — | 64.9% |
Knowledge ERNIE 5.1 leads
ERNIE 5.1: 41.9 (#102), Phi-4: 32.6 (#209)
| Benchmark | ERNIE 5.1 | Phi-4 |
|---|---|---|
| LMArena Expert | 1492 | 1203 |
| GPQA Diamond | — | 56.1% |
| Confabulations | — | 29.4% |
| Vectara Hallucination Rate | — | 3.7% |
| MMLU | — | 84.8% |
Multilingual ERNIE 5.1 leads
ERNIE 5.1: 55.5 (#29), Phi-4: 37.2 (#237)
| Benchmark | ERNIE 5.1 | Phi-4 |
|---|---|---|
| LMArena Non-English | 1454 | 1197 |
| LMArena Chinese | 1508 | 1212 |
| LMArena French | 1488 | 1224 |
| LMArena German | 1470 | 1222 |
| LMArena Japanese | 1422 | 1158 |
| LMArena Korean | 1427 | 1151 |
| LMArena Russian | 1459 | 1209 |
| LMArena Spanish | 1473 | 1234 |
Instruction Following ERNIE 5.1 leads
ERNIE 5.1: 76.7 (#37), Phi-4: 60.4 (#251)
| Benchmark | ERNIE 5.1 | Phi-4 |
|---|---|---|
| LMArena Instruction Following | 1460 | 1201 |
| LiveBench Instruction Following | — | 58.4% |
Long Context ERNIE 5.1 leads
ERNIE 5.1: 44.7 (#59), Phi-4: 36.9 (#226)
| Benchmark | ERNIE 5.1 | Phi-4 |
|---|---|---|
| LMArena Longer Query | 1462 | 1217 |
Writing & Preference ERNIE 5.1 leads
ERNIE 5.1: 65.1 (#52), Phi-4: 40.5 (#244)
| Benchmark | ERNIE 5.1 | Phi-4 |
|---|---|---|
| LMArena Text | 1468 | 1217 |
| LMArena Creative Writing | 1441 | 1182 |
| LMArena Multi-Turn | 1471 | 1206 |
| Short-Story Creative Writing | — | 62.6% |
| LiveBench Language | — | 25.6% |
Frequently asked questions
Is ERNIE 5.1 better than Phi-4?
ERNIE 5.1 is the stronger model overall, scoring 43.8 to 31.2 on the Noometry Index.
Is ERNIE 5.1 or Phi-4 better for coding?
ERNIE 5.1 scores higher on coding benchmarks: 44.0 versus 34.4 in the Noometry coding category.
How many benchmarks do ERNIE 5.1 and Phi-4 share?
17 benchmarks have published results for both models. ERNIE 5.1 has 19 scored results on Noometry and Phi-4 has 37.