ERNIE 4.0 Pro: pricing, context window and benchmarks
ERNIE 4.0 Pro is a multimodal model from Baidu. It offers a 200,000-token context window, costs $12.58 per million input tokens and $59.04 per million output tokens, and averages 77.5 across the index's benchmark suite.
Pricing
ERNIE 4.0 Pro costs $12.58 per million input tokens and $59.04 per million output tokens. A blended 3:1 input-to-output mix works out to about $24.20 per million tokens. Cached input is listed at $1.60 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $12.58 | per 1M tokens |
| Output | $59.04 | per 1M tokens |
| Cached input | $1.60 | per 1M tokens |
| Blended 3:1 | $24.20 | estimate per 1M tokens |
| 1M in + 1M out | $71.62 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 81.7 | 129 | 79th | |
| GPQA Diamond | 71.7 | 225 | 63th | |
| SWE-bench Verified | 65.7 | 199 | 67th | |
| HumanEval | 81.7 | 191 | 68th | |
| MATH-500 | 82.1 | 163 | 73th | |
| AIME 2025 | 75.6 | 147 | 76th | |
| LiveCodeBench | 68.8 | 212 | 65th | |
| IFEval | 86.0 | 155 | 74th | |
| TruthfulQA | 78.5 | 237 | 61th | |
| HellaSwag | 86.5 | 191 | 68th | |
| MMMU | 74.6 | 167 | 72th | |
| Arena Elo | 1418.3 | 156 | 74th |
What ERNIE 4.0 Pro is good for
ERNIE 4.0 Pro is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $12.58 per million input tokens, it sits in the mid-range of the index, and the 200,000-token context window is enough for large files and multi-file diffs.
Strengths
- Large 200,000-token context window for long documents and agent traces.
- Accepts more than text, including image input where the provider supports it.
- Coding benchmarks are a relative strength within its tier.
Trade-offs
- Proprietary licence: you cannot download the weights or self-host the model.
- Premium input pricing at $12.58 per million tokens.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Codestral | Mistral AI | 65536 | $4.19 | 77.6 |
| Hunyuan-Turbo Pro | Tencent | 65536 | $12.00 | 77.6 |
| ERNIE 4.0 Pro | Baidu | 200000 | $12.58 | 77.5 |
| Hunyuan-Turbo Reasoning | Tencent | 128000 | $11.89 | 77.5 |
| Qwen3 Base | Alibaba Cloud | 65536 | $4.33 | 77.4 |
| Reka Core | Reka AI | 32768 | $12.17 | 77.4 |
FAQ
How much does ERNIE 4.0 Pro cost?
ERNIE 4.0 Pro is listed at $12.58 per million input tokens and $59.04 per million output tokens. Cached input is $1.60 per million tokens. A blended 3:1 input-to-output mix is about $24.20 per million tokens.
What is the context window of ERNIE 4.0 Pro?
ERNIE 4.0 Pro has a context window of 200,000 tokens and can emit up to 32,768 output tokens per request in this index.
Is ERNIE 4.0 Pro open weights?
No. ERNIE 4.0 Pro is a proprietary model with a proprietary licence, so you cannot download the weights.
How does ERNIE 4.0 Pro score on benchmarks?
ERNIE 4.0 Pro averages 77.5 across the scored benchmarks in this index, ranking #185 of 600 models. Its strongest relative result is Arena Elo at 1418.3.
What is ERNIE 4.0 Pro best for?
ERNIE 4.0 Pro is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $12.58 per million input tokens, it sits in the mid-range of the index, and the 200,000-token context window is enough for large files and multi-file diffs.