- SimpleQA 74
- ChineseSimpleQA 86
- MMLU-Pro 83.8
- HLE 25.8
- GPQA-Diamond 86.4
- AIME 2025 89.1
- HMMT 2025 79.6
- LiveCodeBench v6 76.2
- HumanEval+ 94.5
- MBPP+ 82.5
- ZebraLogic 96.5
- BBEH 66.6
- IFEval 93.3
- MultiChallenge 66
- Multi-IF 85.6
- TAU2-Bench 78.8
- Mean of ACEBench-en 87.70 / ACEBench-zh 89.60 88.7
- BFCL v4 66.5
- BrowseComp-zh 64.7
- SpreadSheetBench 40.1
- MMMU-Pro 68.6
- MathVista 84.8
- MathVerse 85.1
- MathVision 74.3
- VisualPuzzle 64.8
- VisuLogic 32
- VLMAreBlind 91.4
- ChartQA 87.8
- AI2D 96.9
- DocVQA (val) 95.5
- OCRBench 878 (out of 1000) 87.8
- ChartXiv-RQ (DQ 89.05) 67.1
- SimpleVQA 67.6
- HallusionBench 63.9
- MMStar 75.5
- BLINK 70
- CV-Bench 87.2
- CountBench 96.5
- VideoMME (w sub) 81.3
- Video-MMMU 81.1
- MMVU 72.2
- GenEval 90.1
- VBench Overall (Quality 84.40, Semantic 83.40) 84.2
ERNIE 5.0
ERNIE 5.0 ranks first on 3 of the 41 benchmarks it has scores on.
48
Benchmarks with a score
3
Ranked first, out of five or more models
9
In the top three
Baidu Jan 2026 ERNIE 5.0
Closed weights Flagship
No Epoch Capabilities Index score yet
- MMMU-Pro 68.6% #44 /72
- BFCL V4 66.5% #11 /24
- SimpleVQA 67.6% #10 /23
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Multi-IF | 85.6% | #2 of 28 | Model card | |
| Chinese SimpleQA | 86% | #1 of 13 | Model card | |
| IFEval | 93.3% | #12 of 126 | Model card | |
| SimpleQA | 74% | #8 of 59 | Model card | |
| BIG-Bench Extra Hard | 66.6% | #2 of 11 | Model card | |
| MultiChallenge | 66% | #10 of 36 | Model card | |
| MMLU-Pro | 83.8% | #42 of 151 | Model card | |
| SimpleVQA | 67.6% | #10 of 23 | Model card | |
| Humanity's Last Exam | 25.8% | #83 of 147 | Model card | |
| HallusionBench | 63.9% | #13 of 23 | Model card | |
| Extended NYT Connections | 10.3% | #92 of 97 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| ZebraLogic | 96.5% | #2 of 23 | Model card | |
| GPQA Diamond | 86.4% | #62 of 284 | Model card | |
| AIME 2025 | 89.1% | #56 of 143 | Model card | |
| HMMT 2025 | 79.6% | #34 of 49 | Model card |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HumanEval+ | 94.5% | #1 of 27 | Model card | |
| MBPP+ | 82.5% | #2 of 15 | Model card | |
| LiveCodeBench v6 | 76.2% | #38 of 76 | Model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AceBench | 88.7% | #1 of 3 | Model card | |
| BFCL V4 | 66.5% | #11 of 24 | Model card | |
| tau2-bench | 78.8% | #38 of 69 | Model card | |
| BrowseComp-ZH | 64.7% | #10 of 17 | Model card |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SpreadsheetBench | 40.1% | #6 of 6 | Model card |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AI2D | 96.9% | #1 of 44 | Model card | |
| MathVerse | 85.1% | #2 of 12 | Model card | |
| DocVQA | 95.5% | #7 of 33 | Model card | |
| BLINK | 70% | #6 of 28 | Model card | |
| MMAU | 80.4% | #3 of 14 | Model card | |
| MathVista | 84.8% | #15 of 64 | Model card | |
| CountBench | 96.5% | #5 of 20 | Model card | |
| AISHELL-1 | 0.31% | #1 of 4 | Model card | |
| OCRBench | 87.8% | #9 of 28 | Model card | |
| Clotho-AQA | 73.7% | #1 of 3 | Model card | |
| GenEval | 90.1% | #1 of 3 | Model card | |
| VisualPuzzles | 64.8% | #1 of 3 | Model card | |
| ChartQA | 87.8% | #13 of 37 | Model card | |
| CVBench | 87.2% | #4 of 11 | Model card | |
| MATH-Vision (MathVision) | 74.3% | #21 of 42 | Model card | |
| MMStar | 75.5% | #15 of 30 | Model card | |
| MMVU | 72.2% | #10 of 20 | Model card | |
| Video-MME | 81.3% | #22 of 42 | Model card | |
| MMMU-Pro | 68.6% | #44 of 72 | Model card | |
| Video-MMMU | 81.1% | #20 of 30 | Model card | |
| AISHELL-2 | 2.64% | #2 of 3 | Model card | |
| VisuLogic | 32% | #5 of 7 | Model card | |
| VLMs Are Blind | 91.4% | #6 of 8 | Model card | |
| CharXiv | 67.1% | #52 of 67 | Model card | |
| VBench | 84.2% | #1 of 1 | Model card |
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.
Its system card
Every result in the longer documents the lab published with it. 9 of them aren't in the launch post.
ERNIE 5.0 Technical Report 4 Feb 2026
Capabilities
- SimpleQA 74
- Chinese SimpleQA 86
- MMLU-Pro 83.8
- Humanity's Last Exam 25.8
- GPQA Diamond 86.4
- AIME 2025 89.1
- HMMT 2025 79.6
- LiveCodeBench v6 76.2
- HumanEval+ 94.5
- MBPP+ 82.5
- ZebraLogic 96.5
- BIG-Bench Extra Hard 66.6
- IFEval 93.3
- MultiChallenge 66
- Multi-IF 85.6
- tau2-bench 78.8
- AceBench
- BFCL V4 66.5
- BrowseComp-ZH 64.7
- SpreadsheetBench 40.1
- MMMU-Pro 68.6
- MathVista 84.8
- MathVerse 85.1
- MATH-Vision (MathVision) 74.3
- VisualPuzzles 64.8
- VisuLogic 32
- VLMs Are Blind 91.4
- ChartQA 87.8
- AI2D 96.9
- DocVQA 95.5
- OCRBench 87.8
- CharXiv 67.1
- SimpleVQA 67.6
- HallusionBench 63.9
- MMStar 75.5
- BLINK 70
- CVBench 87.2
- CountBench 96.5
- Video-MME 81.3
- Video-MMMU 81.1
- MMVU 72.2
- GenEval 90.1
- VBench 84.2
- AISHELL-1 0.31
- AISHELL-2 2.64
- WenetSpeech
- LibriSpeech
- FLEURS
- VoiceBench
- MMAU 80.4
- Clotho-AQA 73.7
- Seed-TTS-eval