- MMLU 86.5
- MMLU-Pro 78.4
- C-Eval 90.6
- CMMLU 90.2
- LiveBench 68.9
- ChineseSimpleQA 77.1
- SimpleQA 45.9
- IFEval 88
- Multi-IF 76.6
- Sysbench 69.8
- MATH-500 96.4
- GSM8K 96.6
- CMath 96.7
- OlympicArena 75.3
- AIME'24 54.8
- AIME'25 35.1
- BBH 94.3
- MUSR 69.9
- DROP 91.1
- Zebralogic 58.1
- LiveCodeBench 38.8
- Humaneval+ 92.1
- MBPP+ 81
- FullStackBench 67.1
ERNIE-4.5-300B-A47B
ERNIE-4.5-300B-A47B ranks first on 2 of the 29 benchmarks it has scores on.
30
Benchmarks with a score
2
Ranked first, out of five or more models
6
In the top three
Baidu Jun 2025 ERNIE-4.5-300B-A47B
Open weights Flagship
No Epoch Capabilities Index score yet
- FullStackBench 67.1% #3 /6
- SysBench 69.8% #5 /6
- MMLU-Pro 78.4% #84 /151
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 | |
|---|---|---|---|---|
| BIG-Bench Hard | 94.3% | #1 of 71 | Model card | |
| CMMLU | 90.2% | #3 of 17 | Model card | |
| DROP | 91.1% | #8 of 40 | Model card | |
| SimpleQA | 45.9% | #15 of 59 | Model card | |
| Chinese SimpleQA | 77.1% | #4 of 13 | Model card | |
| Multi-IF | 76.6% | #9 of 28 | Model card | |
| MuSR | 69.9% | #2 of 6 | Model card | |
| MMLU | 86.5% | #44 of 127 | Model card | |
| IFEval | 88% | #46 of 126 | Model card | |
| C-Eval | 90.6% | #15 of 38 | Model card | |
| LLM Confabulation Leaderboard | 16% | #28 of 58 | Official leaderboard | |
| MMLU-Pro | 78.4% | #84 of 151 | Model card | |
| SysBench | 69.8% | #5 of 6 | Model card | |
| BullshitBench | 12.7% | #93 of 107 | Official leaderboard | |
| BullshitBench V2 | 4.00% | #119 of 122 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| GSM8K | 96.6% | #2 of 87 | Model card | |
| CMATH | 96.7% | #1 of 6 | Model card | |
| MATH-500 | 96.4% | #28 of 67 | Model card | |
| AIME 2024 | 54.8% | #72 of 87 | Model card | |
| AIME 2025 | 35.1% | #136 of 143 | Model card | |
| ZebraLogic | 58.1% | #22 of 23 | Model card | |
| OlympicArena | 75.3% | #1 of 1 | Model card |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HumanEval+ | 92.1% | #6 of 27 | Model card | |
| MBPP+ | 81% | #4 of 15 | Model card | |
| FullStackBench | 67.1% | #3 of 6 | Model card | |
| LiveCodeBench | 38.8% | #50 of 73 | Model card | |
| LiveCodeBench v6 | 38.8% | #74 of 76 | Model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SnakeBench | 23 | #99 of 183 | Official leaderboard | |
| LisanBench | 176.3 | #94 of 133 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveBench | 68.9% | #13 of 27 | 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. 1 of them isn't in the launch post.
ERNIE 4.5 Technical Report 29 Jun 2025
Capabilities
- C-Eval 90.6
- CMMLU 90.2
- MMLU 86.5
- MMLU-Pro 78.4
- LiveBench 68.9
- BIG-Bench Hard 94.3
- MuSR 69.9
- DROP 91.1
- ZebraLogic 58.1
- MATH-500 96.4
- GSM8K 96.6
- CMATH 96.7
- OlympicArena 75.3
- AIME 2024 54.8
- AIME 2025 35.1
- LiveCodeBench v6 38.8
- HumanEval+ 92.1
- MBPP+ 81
- FullStackBench 67.1
- IFEval 88
- Multi-IF 76.6
- SysBench 69.8
Hallucination
- SimpleQA 45.9
- Chinese SimpleQA 77.1