ERNIE 4.5 21B A3B PT 8bit by mlx-community

 »  All LLMs  »  mlx-community  »  ERNIE 4.5 21B A3B PT 8bit   URL Share it on

ERNIE 4.5 21B A3B PT 8bit is an open-source language model by mlx-community. Features: 21b LLM, VRAM: 23.2GB, Context: 128K, License: apache-2.0, Quantized, LLM Explorer Score: 0.2.

  8-bit   8bit Base model:baidu/ernie-4.5-21b... Base model:quantized:baidu/ern...   Conversational   Custom code   En   Ernie4.5   Ernie4 5 moe   Mlx   Quantized   Region:us   Safetensors   Sharded   Tensorflow   Zh

ERNIE 4.5 21B A3B PT 8bit Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

ERNIE 4.5 21B A3B PT 8bit Parameters and Internals

LLM NameERNIE 4.5 21B A3B PT 8bit
Repository 🤗https://huggingface.co/mlx-community/ERNIE-4.5-21B-A3B-PT-8bit 
Base Model(s)  baidu/ERNIE-4.5-21B-A3B-PT   baidu/ERNIE-4.5-21B-A3B-PT
Model Size21b
Required VRAM23.2 GB
Updated2026-04-09
Maintainermlx-community
Model Typeernie4_5_moe
Model Files  5.2 GB: 1-of-5   5.4 GB: 2-of-5   5.3 GB: 3-of-5   5.3 GB: 4-of-5   2.0 GB: 5-of-5
Supported Languagesen zh
Quantization Type8bit
Model ArchitectureErnie4_5_MoeForCausalLM
Licenseapache-2.0
Context Length131072
Model Max Length131072
Tokenizer ClassErnie4_5_Tokenizer
Padding Token<unk>
Vocabulary Size103424
Torch Data Typebfloat16

Best Alternatives to ERNIE 4.5 21B A3B PT 8bit

Best Alternatives
Context / RAM
Downloads
Likes
... Claude 4.5 High OPUS Thinking128K / 43.7 GB150
ERNIE 4.5 21B A3B Thinking128K / 43.7 GB630777
ERNIE 4.5 21B A3B PT128K / 44 GB33224168
ERNIE 4.5 21B A3B Base PT128K / 43.9 GB368931
ERNIE 4.5 21B A3B Thinking128K / 43.7 GB1507
ERNIE 4.5 21B A3B PT Bf16128K / 43.7 GB51
Note: green Score (e.g. "73.2") means that the model is better than mlx-community/ERNIE-4.5-21B-A3B-PT-8bit.

Rank the ERNIE 4.5 21B A3B PT 8bit Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

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  

What open-source LLMs or SLMs are you in search of? 52840 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a