Baidu: ERNIE 4.5 VL 424B A47B

By Baidu

Released: 2025-06-30

Visit Website

API

LLM

RAG

Baidu

Paid

New

Baidu Ernie 4 5 Vl 424B A47B features 47B parameters, MoE architecture and 123k-token context window. 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 optimised for efficient scaling and can operate under 4-bit/8-bit quantization. Available at $0.42/1M tokens.

Specifications

AI Evaluation