What are the hardware requirements for ERNIE 4.5 300B A47B FP8 Paddle?

ERNIE 4.5 300B A47B FP8 Paddle requires approximately 219.6 GB of VRAM and supports a context window of 128K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed ERNIE 4.5 300B A47B FP8 Paddle and how large is it?

ERNIE 4.5 300B A47B FP8 Paddle is developed by baidu, a model with 300b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate ERNIE 4.5 300B A47B FP8 Paddle?

ERNIE 4.5 300B A47B FP8 Paddle is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

ERNIE 4.5 300B A47B FP8 Paddle by baidu — VRAM 219.6GB, 128K context

Name: ERNIE 4.5 300B A47B FP8 Paddle
Author: baidu

ERNIE 4.5 300B A47B FP8 Paddle is an open-source language model by baidu. Features: 300b LLM, VRAM: 219.6GB, Context: 128K, License: apache-2.0, LLM Explorer Score: 0.19.

8-bit Conversational En Ernie4.5 Ernie4 5 moe Paddlepaddle Region:us Safetensors Sharded Tensorflow Zh

Model Card on HF 🤗: https://huggingface.co/baidu/ERNIE-4.5-300B-A47B-FP8-Paddle

ERNIE 4.5 300B A47B FP8 Paddle Parameters and Internals

LLM Name	ERNIE 4.5 300B A47B FP8 Paddle
Repository 🤗	https://huggingface.co/baidu/ERNIE-4.5-300B-A47B-FP8-Paddle
Model Size	300b
Required VRAM	219.6 GB
Updated	2026-08-08
Maintainer	baidu
Model Type	ernie4_5_moe
Model Files	5.0 GB: 1-of-61 5.0 GB: 2-of-61 5.0 GB: 3-of-61 5.0 GB: 4-of-61 5.0 GB: 5-of-61 5.0 GB: 6-of-61 5.0 GB: 7-of-61 4.8 GB: 8-of-61 5.0 GB: 9-of-61 5.0 GB: 10-of-61 5.0 GB: 11-of-61 5.0 GB: 12-of-61 5.0 GB: 13-of-61 5.0 GB: 14-of-61 4.9 GB: 15-of-61 4.9 GB: 16-of-61 5.0 GB: 17-of-61 5.0 GB: 18-of-61 5.0 GB: 19-of-61 5.0 GB: 20-of-61 5.0 GB: 21-of-61 5.0 GB: 22-of-61 5.0 GB: 23-of-61 5.0 GB: 24-of-61 5.0 GB: 25-of-61 5.0 GB: 26-of-61 5.0 GB: 27-of-61 5.0 GB: 28-of-61 5.0 GB: 29-of-61 5.0 GB: 30-of-61 5.0 GB: 31-of-61 5.0 GB: 32-of-61 5.0 GB: 33-of-61 5.0 GB: 34-of-61 5.0 GB: 35-of-61 5.0 GB: 36-of-61 5.0 GB: 37-of-61 5.0 GB: 38-of-61 5.0 GB: 39-of-61 5.0 GB: 40-of-61 5.0 GB: 41-of-61 5.0 GB: 42-of-61 5.0 GB: 43-of-61 5.0 GB: 44-of-61
Supported Languages	en zh
Model Architecture	Ernie4_5_MoeForCausalLM
License	apache-2.0
Context Length	131072
Model Max Length	131072
Vocabulary Size	103424

Rank the ERNIE 4.5 300B A47B FP8 Paddle capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to ERNIE 4.5 300B A47B FP8 Paddle

Best Alternatives	Context / RAM	Downloads	Likes
ERNIE 4.5 300B A47B PT	128K / 205.9 GB	2108	78
ERNIE 4.5 300B A47B Paddle	128K / 210.8 GB	254	16
ERNIE 4.5 300B A47B Base PT	128K / 210.8 GB	69	41
...4.5 300B A47B 2Bits TP2 Paddle	128K / GB	32	11
...NIE 4.5 300B A47B 2Bits Paddle	128K / 93.9 GB	37	24
...RNIE 4.5 300B A47B Base Paddle	128K / 215.7 GB	35	22
...4.5 300B A47B 2Bits TP4 Paddle	128K / GB	30	5
....5 300B A47B W4A8C8 TP4 Paddle	128K / GB	28	16
ERNIE 4.5 300B A47B PT 4bit	128K / 169 GB	47	2

Note: green Score (e.g. "73.2") means that the model is better than baidu/ERNIE-4.5-300B-A47B-FP8-Paddle.