ERNIE 4.5 Turbo: pricing, context window and benchmarks
ERNIE 4.5 Turbo is a multimodal model from Baidu. It offers a 128,000-token context window, costs $13.75 per million input tokens and $54.36 per million output tokens, and averages 79.7 across the index's benchmark suite.
Pricing
ERNIE 4.5 Turbo costs $13.75 per million input tokens and $54.36 per million output tokens. A blended 3:1 input-to-output mix works out to about $23.90 per million tokens. Cached input is listed at $1.49 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $13.75 | per 1M tokens |
| Output | $54.36 | per 1M tokens |
| Cached input | $1.49 | per 1M tokens |
| Blended 3:1 | $23.90 | estimate per 1M tokens |
| 1M in + 1M out | $68.11 | 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 | 84.0 | 80 | 87th | |
| GPQA Diamond | 73.0 | 198 | 67th | |
| SWE-bench Verified | 67.5 | 156 | 74th | |
| HumanEval | 86.4 | 98 | 84th | |
| MATH-500 | 86.4 | 86 | 86th | |
| AIME 2025 | 80.7 | 79 | 87th | |
| LiveCodeBench | 70.5 | 179 | 70th | |
| IFEval | 85.2 | 183 | 70th | |
| TruthfulQA | 82.5 | 117 | 81th | |
| HellaSwag | 87.0 | 177 | 71th | |
| MMMU | 73.7 | 199 | 67th | |
| Arena Elo | 1419.8 | 151 | 75th |
What ERNIE 4.5 Turbo is good for
ERNIE 4.5 Turbo 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 $13.75 per million input tokens, it sits in the mid-range of the index, and the 128,000-token context window is enough for large files and multi-file diffs.
Strengths
- Large 128,000-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 $13.75 per million tokens.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Magistral | Mistral AI | 200000 | $4.75 | 79.8 |
| Qwen3 Instruct | Alibaba Cloud | 1000000 | $5.06 | 79.8 |
| Doubao 1.5 Pro | ByteDance | 200000 | $13.65 | 79.7 |
| ERNIE 4.5 Turbo | Baidu | 128000 | $13.75 | 79.7 |
| Llama 4 Maverick Chat | Meta AI | 128000 | $5.09 | 79.7 |
| Sonar Pro Pro | Perplexity | 200000 | $13.59 | 79.7 |
FAQ
How much does ERNIE 4.5 Turbo cost?
ERNIE 4.5 Turbo is listed at $13.75 per million input tokens and $54.36 per million output tokens. Cached input is $1.49 per million tokens. A blended 3:1 input-to-output mix is about $23.90 per million tokens.
What is the context window of ERNIE 4.5 Turbo?
ERNIE 4.5 Turbo has a context window of 128,000 tokens and can emit up to 32,768 output tokens per request in this index.
Is ERNIE 4.5 Turbo open weights?
No. ERNIE 4.5 Turbo is a proprietary model with a proprietary licence, so you cannot download the weights.
How does ERNIE 4.5 Turbo score on benchmarks?
ERNIE 4.5 Turbo averages 79.7 across the scored benchmarks in this index, ranking #129 of 600 models. Its strongest relative result is Arena Elo at 1419.8.
What is ERNIE 4.5 Turbo best for?
ERNIE 4.5 Turbo 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 $13.75 per million input tokens, it sits in the mid-range of the index, and the 128,000-token context window is enough for large files and multi-file diffs.