ERNIE 4.5 21B A3B APEX needs about 12.2 GB of memory at its main file (a 11.2 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 2 quants, from 11.2 GB to 15.3 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 1.9 GB.
ERNIE 4.5 21B A3B APEX is a 21.8B-parameter Ernie model in GGUF format, quantized by Myric. Its trained context length is 128K tokens. The repository has 2 quant files, from 11.2 GB to 15.3 GB.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 11.2 GB | 12.2 GB | Too big | Get file → | |
| — | 15.3 GB | 16.3 GB | Too big | Get file → |
What GPU do you need for ERNIE 4.5 21B A3B APEX?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: the best quant that fits is the main file (11.2 GB file, about 12.2 GB in use).
- 24 GB VRAM: the best quant that fits is the main file (15.3 GB file, about 16.3 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about ERNIE 4.5 21B A3B APEX
How much VRAM does ERNIE 4.5 21B A3B APEX need?
About 12.2 GB at its main file with an 8K context window. That is the 11.2 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run ERNIE 4.5 21B A3B APEX on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 12.2 GB, so you would need a larger GPU or CPU offloading.
What is the file size of ERNIE 4.5 21B A3B APEX GGUF?
It depends on the quant: file is 11.2 GB, file is 15.3 GB.
What context length does ERNIE 4.5 21B A3B APEX support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.5 GB at 8K and 1.9 GB at 32K.
Where can I download ERNIE 4.5 21B A3B APEX GGUF?
The files are on Hugging Face in the Myric/ERNIE-4.5-21B-A3B-APEX-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does ERNIE 4.5 21B A3B APEX use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.