# ERNIE 4.5 21B A3B

A sophisticated text-based Mixture-of-Experts (MoE) model featuring 21B total parameters with 3B activated per token, delivering exceptional multimodal understanding and generation through heterogeneous MoE structures and modality-isolated routing. Supporting an extensive 131K token context length, the model achieves efficient inference via multi-expert parallel collaboration and quantization, while advanced post-training techniques including SFT, DPO, and UPO ensure optimized performance across diverse applications with specialized routing and balancing losses for superior task handling.

## Model Information

- **Organization**: [baidu](/llm.txt)
- **Slug**: ernie-4-5-21b-a3b
- **Available at Providers**: 9

## Providers

| Provider | Name | $ Input (per 1M) | $ Output (per 1M) | Free | Link |
|----------|------|-----------------|------------------|------|------|
| [AIMLAPI](/llm/aimlapi.txt) | Ernie 4.5 21B A3B |  |  |  |  |
| [Novita AI](/llm/novita.txt) | ernie-4.5-21B-a3b | 0.07 | 0.28 |  |  |
| [OpenRouter](/llm/openrouter.txt) | ERNIE 4.5 21B A3B | 0.07 | 0.28 |  | [View](https://openrouter.ai/baidu/ernie-4.5-21b-a3b) |
| [ValorGPT](/llm/valorgpt.txt) | BA |  |  |  | [View](https://www.valorgpt.com/models/baidu-ernie-4.5-21b-a3b) |
| [Yupp](/llm/yupp.txt) | ERNIE 4.5 21B A3B (Novita) |  |  |  |  |
| [Yupp](/llm/yupp.txt) | ERNIE 4.5 21B A3B (OpenRouter) |  |  |  |  |
| [LangDB](/llm/langdb.txt) | ernie-4.5-21b-a3b |  |  |  | [View](https://langdb.ai/app/models) |
| [Kilo Code](/llm/kilocode.txt) | Baidu: ERNIE 4.5 21B A3B | 0.07 | 0.28 |  |  |
| [Baidu AI Studio](/llm/baidu.txt) |  |  |  |  | [View](https://aistudio.baidu.com) |

---

[← Back to all providers](/llm.txt)