ERNIE 4.5 VL 424B A47B

About ERNIE 4.5 VL 424B A47B

ERNIE-4.5-VL-424B-A47B is a multimodal Mixture-of-Experts (MoE) model from Baidu’s ERNIE 4.5 series, featuring 424B total parameters with 47B active per token. It is trained jointly on text and image data using a heterogeneous MoE architecture and modality-isolated routing to enable high-fidelity cross-modal reasoning, image understanding, and long-context generation (up to 131k tokens). Fine-tuned with techniques like SFT, DPO, UPO, and RLVR, this model supports both “thinking” and non-thinking inference modes. Designed for vision-language tasks in English and Chinese, it is optimized for efficient scaling and can operate under 4-bit/8-bit quantization.

Model Specifications

💰 Real-World Cost Examples

Estimated monthly costs for common use cases

Baidu Model Lineup

Compare all models from Baidu to find the best fit

Similar Models from Other Providers

Cross-brand alternatives with similar capabilities

Qwen2.5 Coder 32B Instruct

Alibaba Qwen

Input: $0.66

Output: $1.00

Context: 33k

Perceptron Mk1

Perceptron

Input: $0.15

Output: $1.50

Context: 33k

Qwen3 VL 30B A3B Thinking

Alibaba Qwen

Input: $0.13

Output: $1.56

Context: 262k

Llama 3.1 Euryale 70B v2.2

Sao10K

Input: $0.85

Output: $0.85

Context: 131k

💡 Cheaper Alternatives

Same Brand (Baidu)

Cross Brand

🚀 Quick Start

Get started with ERNIE 4.5 VL 424B A47B API

from openai import OpenAI

client = OpenAI(
    base_url="https://api.provider.com/v1",
    api_key="YOUR_API_KEY"
)

response = client.chat.completions.create(
    model="baidu/ernie-4.5-vl-424b-a47b",
    messages=[
        {"role": "user", "content": "Hello!"}
    ]
)
print(response.choices[0].message.content)

Continue Your Decision