Back to Models
BaiduOpen Source
ERNIE-4.5-VL-28B-A3B Multimodal MoE Chat Model
ERNIE-4.5-VL-28B-A3B is a multi-modal foundation model developed by Baidu. It adopts a MoE architecture, featuring a vast parameter count and a long 131K context window.
Parameters
280.0B
Context Window
131K
License
Apache 2.0
Release Date
2025-08-20
API Pricing
API pricing for this model is not yet available
Strengths
- ・Large parameter count of 280B
- ・Supports long 131K context
- ・Uses efficient MoE architecture
Weaknesses
- ・High computational load from model size
- ・Intensive resource consumption
- ・Requires advanced hardware
Use Cases
- ・Complex multimodal analysis
- ・Understanding and processing long contexts
- ・Advanced vision-language integrated chat
Deep Analysis
Parameters
28B total, 3B activated
Context Window
128K tokens
Pricing
$0.14/$0.56 per 1M tokens
License
Apache 2.0
Release Date
August 12, 2025
Strengths
- ・Efficient MoE with 3B activated params
- ・Thinking/non-thinking modes
- ・Vision capabilities
- ・Affordable
Weaknesses
- ・Limited to 30K on APIs
- ・No video/audio input
- ・Lower scores vs larger models
Competitor Comparison
| Model |
|---|
| Qwen3-VL |
| Llama 4 Scout |
ERNIE-4.5-VL-28B-A3B is a multimodal MoE model with 28B/3B activated params. Heterogeneous MoE with modality-isolated routing.
Sources
Analysis generated: 2026-05-24