baidu/ERNIE-4.5-300B-A47B

by Baidu

ERNIE-4.5-300B-A47B is a large language model developed by Baidu based on a Mixture of Experts (MoE) architecture. The model has a total of 300 billion parameters, but only activates 47 billion parameters per token during inference, which balances strong performance with computational efficiency. As one of the core models in the ERNIE 4.5 series, it demonstrates outstanding capabilities in tasks such as text understanding, generation, reasoning, and programming. The model employs an innovative multimodal heterogeneous MoE pretraining approach, leveraging joint training of textual and visual modalities to effectively enhance the model’s overall abilities, particularly excelling in instruction following and world knowledge memorization. Baidu has open-sourced this model along with other models in the series, aiming to promote the research and application of AI technology.

API Pricing

Input$0.32 / 1M tokens
Output$1.28 / 1M tokens

Specifications

Modalitiestext, image
Featurestool calling, function calling, structured outputs

FAQ

What is baidu/ERNIE-4.5-300B-A47B?

ERNIE-4.5-300B-A47B is a large language model developed by Baidu based on a Mixture of Experts (MoE) architecture. The model has a total of 300 billion parameters, but only activates 47 billion parameters per token during inference, which balances strong performance with computational efficiency. As one of the core models in the ERNIE 4.5 series, it demonstrates outstanding capabilities in tasks such as text understanding, generation, reasoning, and programming. The model employs an innovative multimodal heterogeneous MoE pretraining approach, leveraging joint training of textual and visual modalities to effectively enhance the model’s overall abilities, particularly excelling in instruction following and world knowledge memorization. Baidu has open-sourced this model along with other models in the series, aiming to promote the research and application of AI technology.

How much does baidu/ERNIE-4.5-300B-A47B cost?

On AIHubMix, baidu/ERNIE-4.5-300B-A47B costs $0.32 per million input tokens and $1.28 per million output tokens.

What modalities does baidu/ERNIE-4.5-300B-A47B support?

baidu/ERNIE-4.5-300B-A47B accepts text and image input.

What features does baidu/ERNIE-4.5-300B-A47B support?

baidu/ERNIE-4.5-300B-A47B supports tool calling, function calling and structured outputs. Per-protocol parameter support is listed in the capability table on this page.

How do I call baidu/ERNIE-4.5-300B-A47B via API?

baidu/ERNIE-4.5-300B-A47B is available through the AIHubMix unified API. The API is OpenAI-compatible: point your OpenAI SDK at https://aihubmix.com/v1, use your AIHubMix API key, and set the model name to baidu/ERNIE-4.5-300B-A47B — no other code changes needed.

Who develops baidu/ERNIE-4.5-300B-A47B?

baidu/ERNIE-4.5-300B-A47B is developed by Baidu. AIHubMix aggregates it alongside models from other providers behind one API and one bill.

More from Baidu

ernie-5.1

by Baidu

ERNIE 5.1 is the latest model in the Wenxin series, with comprehensive upgrades to its…

$0.56/1M in · $2.54/1M out
119,000 tokens context

ernie-5.0

by Baidu

ERNIE 5.0 is the next-generation natively multimodal foundation model in the ERNIE…

$0.82/1M in · $3.29/1M out
119,000 tokens context

ernie-image-turbo

by Baidu

The Ernie-image-Turbo model is an 8-step distilled version of the Ernie-image model, also…

$2/1M in

musesteamer-air-image

by Baidu

musesteamer-air-image is a text-to-image model developed by the Baidu Search team aimed…

$2/1M in

qianfan-ocr

by Baidu

Qianfan-OCR-Fast is a multimodal large model specialized for OCR, trained primarily on…

$0.06/1M in · $0.25/1M out
32,000 tokens context

qianfan-ocr-fast

by Baidu

Qianfan-OCR-Fast is a multimodal large model specialized for OCR, trained primarily on…

$0.66/1M in · $2.74/1M out
32,000 tokens context

Use baidu/ERNIE-4.5-300B-A47B via the AIHubMix unified API — one interface for every major LLM.