Baidu: ERNIE 4.5 VL 424B A47B

Baidu: ERNIE 4.5 VL 424B A47B

baidu · Released Jun 30, 2025
32
Our Score

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.

$0.42 / 1M Input Price
$1.25 / 1M Output Price
123,000 tokens Context Window
16,000 tokens Max Output
424B Parameters

Capabilities

Vision

Architecture

ModalityText + Image → Text
TokenizerOther
Parameters424B

Model Information

OpenRouter ID baidu/ernie-4.5-vl-424b-a47b
Providerbaidu
Release Date June 30, 2025
Context Length123,000 tokens
Max Completion16,000 tokens
Status Active

Pricing

Token Type Cost per 1M tokens Cost per 1K tokens
Input $0.42 $0.000420
Output $1.25 $0.001250