ERNIE-4.5-21B-A3B-Base-PT: benchmark scores, price and ranking
ERNIE-4.5-21B-A3B-Base-PT from baidu ranks #384 of 743 LLMs on the LLMs Tiger aggregate benchmark score (48.1), based on 1 benchmark results captured up to 2026-10-13.
- Organisation
- baidu
- Type
- Open-weight / local model
- Open weights
- Yes
- Aggregate rank
- #384 of 743 (score 48.1)
Across the 1 benchmark where at least ten models are scored, ERNIE-4.5-21B-A3B-Base-PT is in the top 10% on 0 and in the top half on 1.
ERNIE-4.5-21B-A3B-Base-PT benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| MMLU-Pro | 56.7% | #66 of 136 | HuggingFace MMLU-Pro leaderboard | 2026-10-13 |
About ERNIE-4.5-21B-A3B-Base-PT
ERNIE-4.5-21B-A3B-Base-PT is an open-weight model that can run on your own hardware from baidu. No API price is listed for it yet. We hold 1 benchmark result for it across 1 area, and none is in the top tenth of its board.
Where ERNIE-4.5-21B-A3B-Base-PT stands
On our aggregate score, ERNIE-4.5-21B-A3B-Base-PT is #384 of 743 ranked models, around the middle of the table. Its score is 48.1; the scale is centred on 50, so it sits 1.9 points below the typical model. That beats 48% of the ranked field. Among the 3 ranked baidu models it is #3. The next one up is ERNIE-4.5-300B-A47B-Base-PT at 49.8.
ERNIE-4.5-21B-A3B-Base-PT, benchmark by benchmark
MMLU-Pro puts ERNIE-4.5-21B-A3B-Base-PT at 56.7%, #66 of 136. That beats 51% of the models tested, but 31.4 points short of Darwin-180B-RSI (88.1%). The results either side are internlm3-8b-instruct (57.6%) and granite-4.1-8b (56.0%). It covers 57-subject knowledge + reasoning. The board median is 55.3%, so ERNIE-4.5-21B-A3B-Base-PT is 1.4 points above it. Source: HuggingFace MMLU-Pro leaderboard, captured 2026-10-13.
Reading these numbers
The 1 score on this page comes from 1 source: HuggingFace MMLU-Pro leaderboard (1). No official leaderboard supplied any of them, so treat them as third-party measurements. With 1 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #381 llama3-70b-8192 (48.1)
- #382 gemini-1.5-pro-001-feb24 (48.1)
- #383 Codex + GPT-5.6 Luna (48.1)
- #385 Llama-3.2-11B-Vision-Instruct (48.0)
- #386 Gemini 3.5 Flash-Lite (High) (48.0)
- #387 DeepSeek-V3.1 (48.0)
More from baidu
- ERNIE-4.5-300B-A47B-PT (#150, 51.0)
- ERNIE-4.5-300B-A47B-Base-PT (#234, 49.8)
Compare it on the live leaderboard → · How the aggregate score works