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
Model Card on HF 🤗: https://huggingface.co/mlx-community/ERNIE-4.5-21B-A3B-PT-8bit
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").
What is the LLM Explorer Rank (Score)
ERNIE 4.5 21B A3B PT 8bit Parameters and Internals
| LLM Name | ERNIE 4.5 21B A3B PT 8bit |
| Repository 🤗 | https://huggingface.co/mlx-community/ERNIE-4.5-21B-A3B-PT-8bit |
| Base Model(s) | |
| Model Size | 21b |
| Required VRAM | 23.2 GB |
| Updated | 2025-08-26 |
| Maintainer | mlx-community |
| Model Type | ernie4_5_moe |
| Model Files | |
| Supported Languages | en zh |
| Quantization Type | 8bit |
| Model Architecture | Ernie4_5_MoeForCausalLM |
| License | apache-2.0 |
| Context Length | 131072 |
| Model Max Length | 131072 |
| Tokenizer Class | Ernie4_5_Tokenizer |
| Padding Token | <unk> |
| Vocabulary Size | 103424 |
| Torch Data Type | bfloat16 |
Best Alternatives to ERNIE 4.5 21B A3B PT 8bit
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| ERNIE 4.5 21B A3B PT | 128K / 44 GB | 150281 | 132 |
| ERNIE 4.5 21B A3B PT Bf16 | 128K / 43.7 GB | 38 | 1 |
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! 🌟
What open-source LLMs or SLMs are you in search of? 50900 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124