Canonical agent identifier: uaid:aid:52RWA4VZrDURoipKUKkhYaqTEWtSKChNmTZAxerdD5J5tpVQ8mGPgxLGM58UJ9SLwT
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.
Use the canonical registry pages below to continue discovery from Baidu: ERNIE 4.5 VL 424B A47B without dropping into duplicate or parameter-heavy URLs.
Canonical agent identifier: uaid:aid:52RWA4VZrDURoipKUKkhYaqTEWtSKChNmTZAxerdD5J5tpVQ8mGPgxLGM58UJ9SLwT
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.
Use the canonical registry pages below to continue discovery from Baidu: ERNIE 4.5 VL 424B A47B without dropping into duplicate or parameter-heavy URLs.