# ERNIE 5.1 vs o4-mini

> ERNIE 5.1 is the stronger model overall, scoring 43.1 to 41.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/ernie-5-1-vs-o4-mini
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. ERNIE 5.1 scores higher in 4 categories and o4-mini in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where ERNIE 5.1 leads 65.3 to 54.2.

## Snapshot

| | ERNIE 5.1 | o4-mini |
|---|---|---|
| Provider | Baidu | OpenAI |
| Noometry Index | 43.1 | 41.6 |
| Rank | 87 | 119 |
| Context | — | 200K |
| Input $/M | — | $1.10 |
| Output $/M | — | $4.40 |
| Weights | Proprietary | Proprietary |

## Coding

- ERNIE 5.1: 43.4 (#78)
- o4-mini: 41.6 (#110)

| Benchmark | ERNIE 5.1 | o4-mini |
|---|---|---|
| LMArena Coding | 1488 | 1369 |
| SWE-bench Verified (bash only) | — | 45% |
| Aider Polyglot | — | 72% |
| GSO | — | 3.6% |
| WeirdML | — | 52.6% |
| CadEval | — | 62% |
| ALE-Bench | — | 826.17 |
| AlgoTune | — | 1.72 |

## Agentic & Tool Use

- ERNIE 5.1: —
- o4-mini: 32.3 (#57)

| Benchmark | ERNIE 5.1 | o4-mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 53.2% |
| GDPval | — | 25.3% |
| LMArena Search | 1227 | — |
| METR Time Horizons | — | 63.9% |

## Reasoning

- ERNIE 5.1: 20.9 (#209)
- o4-mini: 24.6 (#143)

| Benchmark | ERNIE 5.1 | o4-mini |
|---|---|---|
| LMArena Hard Prompts | 1481 | 1351 |
| ARC-AGI-2 | — | 6.1% |
| SimpleBench | — | 38.7% |
| Kagi LLM Benchmark | — | 67.6% |
| NYT Connections (extended) | 23.4% | — |
| ARC-AGI-1 | — | 58.7% |
| CritPt | — | 0.6% |
| Chess Puzzles | — | 26% |
| EnigmaEval | — | 9.2% |
| Mystery Game Puzzles | — | 5% |
| DTBench | — | 77.6% |
| LMCA | — | 26.5% |
| Epoch Capabilities Index | — | 145.63 |
| ForecastBench | — | 61.8 |

## Math

- ERNIE 5.1: 37.5 (#96)
- o4-mini: 40.2 (#79)

| Benchmark | ERNIE 5.1 | o4-mini |
|---|---|---|
| LMArena Math | 1481 | 1390 |
| FrontierMath (Tiers 1-3) | — | 36.1% |
| FrontierMath Tier 4 | — | 4.9% |
| OTIS Mock AIME 2024-2025 | — | 81.7% |
| Omni-MATH | — | 72% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 24.8% |
| FrontierMath Tier 4 (v1) | — | 6.3% |

## Knowledge

- ERNIE 5.1: 41.2 (#103)
- o4-mini: 43.5 (#83)

| Benchmark | ERNIE 5.1 | o4-mini |
|---|---|---|
| LMArena Expert | 1491 | 1344 |
| GPQA Diamond | — | 79.6% |
| Humanity's Last Exam | — | 18.1% |
| SimpleQA Verified | — | 19.6% |
| MMLU-Pro | — | 82% |
| Confabulations | — | 15.8% |
| Vectara Hallucination Rate | — | 18.6% |
| GPQA (HELM) | — | 73.5% |

## Multimodal

- ERNIE 5.1: —
- o4-mini: 39.5 (#46)

| Benchmark | ERNIE 5.1 | o4-mini |
|---|---|---|
| LMArena Vision | — | 1194 |
| GeoBench | — | 64% |
| VPCT | — | 57.5% |

## Multilingual

- ERNIE 5.1: 55.7 (#28)
- o4-mini: 47.2 (#149)

| Benchmark | ERNIE 5.1 | o4-mini |
|---|---|---|
| LMArena Non-English | 1455 | 1336 |
| LMArena Chinese | 1507 | 1355 |
| LMArena French | 1486 | 1362 |
| LMArena German | 1470 | 1336 |
| LMArena Japanese | 1422 | 1308 |
| LMArena Korean | 1428 | 1313 |
| LMArena Russian | 1459 | 1334 |
| LMArena Spanish | 1472 | 1346 |

## Instruction Following

- ERNIE 5.1: 76.9 (#35)
- o4-mini: 75.4 (#63)

| Benchmark | ERNIE 5.1 | o4-mini |
|---|---|---|
| LMArena Instruction Following | 1460 | 1321 |
| IFEval | — | 92.8% |

## Long Context

- ERNIE 5.1: 44.7 (#56)
- o4-mini: 45.5 (#31)

| Benchmark | ERNIE 5.1 | o4-mini |
|---|---|---|
| LMArena Longer Query | 1462 | 1315 |
| Fiction.LiveBench | — | 77.8% |

## Writing & Preference

- ERNIE 5.1: 65.3 (#50)
- o4-mini: 54.2 (#148)

| Benchmark | ERNIE 5.1 | o4-mini |
|---|---|---|
| LMArena Text | 1468 | 1353 |
| LMArena Creative Writing | 1440 | 1294 |
| LMArena Multi-Turn | 1471 | 1350 |
| Short-Story Creative Writing | — | 75% |
| WildBench | — | 85.4% |

## FAQ

### Is ERNIE 5.1 better than o4-mini?

ERNIE 5.1 is the stronger model overall, scoring 43.1 to 41.6 on the Noometry Index.

### Is ERNIE 5.1 or o4-mini better for coding?

ERNIE 5.1 scores higher on coding benchmarks: 43.4 versus 41.6 in the Noometry coding category.

### How many benchmarks do ERNIE 5.1 and o4-mini share?

17 benchmarks have published results for both models. ERNIE 5.1 has 19 scored results on Noometry and o4-mini has 60.