ERNIE 4.0 Nano: pricing, context window and benchmarks
ERNIE 4.0 Nano is a mid-range model from Baidu. It offers a 32,768-token context window, costs $4.68 per million input tokens and $22.77 per million output tokens, and averages 62.1 across the index's benchmark suite.
Pricing
ERNIE 4.0 Nano costs $4.68 per million input tokens and $22.77 per million output tokens. A blended 3:1 input-to-output mix works out to about $9.20 per million tokens. Cached input is listed at $0.45 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $4.68 | per 1M tokens |
| Output | $22.77 | per 1M tokens |
| Cached input | $0.45 | per 1M tokens |
| Blended 3:1 | $9.20 | estimate per 1M tokens |
| 1M in + 1M out | $27.45 | 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 | 65.6 | 442 | 26th | |
| GPQA Diamond | 57.1 | 439 | 27th | |
| SWE-bench Verified | 47.0 | 446 | 26th | |
| HumanEval | 67.0 | 422 | 30th | |
| MATH-500 | 63.4 | 431 | 28th | |
| AIME 2025 | 53.8 | 426 | 29th | |
| LiveCodeBench | 53.3 | 415 | 31th | |
| IFEval | 71.2 | 438 | 27th | |
| TruthfulQA | 68.3 | 410 | 32th | |
| HellaSwag | 74.8 | 444 | 26th | |
| MMMU | 61.6 | 414 | 31th | |
| Arena Elo | 1288.3 | 432 | 28th |
What ERNIE 4.0 Nano is good for
ERNIE 4.0 Nano is a general-purpose model in the index. It handles text generation, summarization and question answering, and it can be a sensible default when you need a balance of quality, price and context. At $4.68 per million input tokens with a 32,768-token window, check its benchmark profile against the specific task before committing.
Strengths
- Balanced specification profile with no single dominant strength.
Trade-offs
- Proprietary licence: you cannot download the weights or self-host the model.
- Average benchmark score is below the index median; test it on your own task.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Kimi K1.5 8B | Moonshot AI | 8192 | $1.54 | 62.2 |
| ERNIE 4.0 Nano | Baidu | 32768 | $4.68 | 62.1 |
| SmolLM3 Chat | Hugging Face | 16384 | $1.61 | 62.1 |
| Llama 4 Maverick 7B | Meta AI | 4096 | $1.52 | 61.9 |
| Sonar Nano | Perplexity | 4096 | $4.37 | 61.9 |
| StableLM 2 Base | Stability AI | 8192 | $1.49 | 61.9 |
FAQ
How much does ERNIE 4.0 Nano cost?
ERNIE 4.0 Nano is listed at $4.68 per million input tokens and $22.77 per million output tokens. Cached input is $0.45 per million tokens. A blended 3:1 input-to-output mix is about $9.20 per million tokens.
What is the context window of ERNIE 4.0 Nano?
ERNIE 4.0 Nano has a context window of 32,768 tokens and can emit up to 32,768 output tokens per request in this index.
Is ERNIE 4.0 Nano open weights?
No. ERNIE 4.0 Nano is a proprietary model with a proprietary licence, so you cannot download the weights.
How does ERNIE 4.0 Nano score on benchmarks?
ERNIE 4.0 Nano averages 62.1 across the scored benchmarks in this index, ranking #427 of 600 models. Its strongest relative result is Arena Elo at 1288.3.
What is ERNIE 4.0 Nano best for?
ERNIE 4.0 Nano is a general-purpose model in the index. It handles text generation, summarization and question answering, and it can be a sensible default when you need a balance of quality, price and context. At $4.68 per million input tokens with a 32,768-token window, check its benchmark profile against the specific task before committing.