ERNIE 4.5 Reasoning: pricing, context window and benchmarks
ERNIE 4.5 Reasoning is a multimodal model from Baidu. It offers a 1,000,000-token context window, costs $14.36 per million input tokens and $78.75 per million output tokens, and averages 79.8 across the index's benchmark suite.
Pricing
ERNIE 4.5 Reasoning costs $14.36 per million input tokens and $78.75 per million output tokens. A blended 3:1 input-to-output mix works out to about $30.46 per million tokens. Cached input is listed at $1.90 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $14.36 | per 1M tokens |
| Output | $78.75 | per 1M tokens |
| Cached input | $1.90 | per 1M tokens |
| Blended 3:1 | $30.46 | estimate per 1M tokens |
| 1M in + 1M out | $93.11 | 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 | 80.6 | 162 | 73th | |
| GPQA Diamond | 75.7 | 129 | 79th | |
| SWE-bench Verified | 69.8 | 114 | 81th | |
| HumanEval | 85.0 | 116 | 81th | |
| MATH-500 | 82.5 | 154 | 74th | |
| AIME 2025 | 78.1 | 114 | 81th | |
| LiveCodeBench | 71.9 | 148 | 75th | |
| IFEval | 88.7 | 88 | 85th | |
| TruthfulQA | 82.9 | 108 | 82th | |
| HellaSwag | 88.0 | 140 | 77th | |
| MMMU | 74.4 | 171 | 72th | |
| Arena Elo | 1421.5 | 141 | 77th |
What ERNIE 4.5 Reasoning is good for
ERNIE 4.5 Reasoning 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 $14.36 per million input tokens, it sits in the mid-range of the index, and the 1,000,000-token context window is enough for large files and multi-file diffs.
Strengths
- Large 1,000,000-token context window for long documents and agent traces.
- Accepts more than text, including image input where the provider supports it.
- Reasoning-oriented profile, useful for multi-step problems and structured analysis.
- 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 $14.36 per million tokens.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Command A Turbo | Cohere | 32768 | $13.53 | 79.8 |
| ERNIE 4.5 Reasoning | Baidu | 1000000 | $14.36 | 79.8 |
| GLM-4.6 Chat | Zhipu AI | 400000 | $4.89 | 79.8 |
| 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 |
FAQ
How much does ERNIE 4.5 Reasoning cost?
ERNIE 4.5 Reasoning is listed at $14.36 per million input tokens and $78.75 per million output tokens. Cached input is $1.90 per million tokens. A blended 3:1 input-to-output mix is about $30.46 per million tokens.
What is the context window of ERNIE 4.5 Reasoning?
ERNIE 4.5 Reasoning has a context window of 1,000,000 tokens and can emit up to 8,192 output tokens per request in this index.
Is ERNIE 4.5 Reasoning open weights?
No. ERNIE 4.5 Reasoning is a proprietary model with a proprietary licence, so you cannot download the weights.
How does ERNIE 4.5 Reasoning score on benchmarks?
ERNIE 4.5 Reasoning averages 79.8 across the scored benchmarks in this index, ranking #124 of 600 models. Its strongest relative result is Arena Elo at 1421.5.
What is ERNIE 4.5 Reasoning best for?
ERNIE 4.5 Reasoning 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 $14.36 per million input tokens, it sits in the mid-range of the index, and the 1,000,000-token context window is enough for large files and multi-file diffs.