baidu.ERNIE 4.5 300B A47B PT needs about 182.5 GB of memory at Q4_K_M (a 180.2 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 12 quants, from 0.3 GB to 599.0 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 7.2 GB.
ERNIE‑4.5 300B A47B PT – a large MoE language model
ERNIE‑4.5 300B A47B PT is a text‑only mixture‑of‑experts model with 300 billion total parameters and 47 billion active parameters. It was trained on 131 072 context tokens and is released under the Apache‑2.0 license.
Good at
- Large‑scale text generation
- Complex reasoning tasks
- Multimodal pre‑training background
Limits
- Requires high VRAM for full precision
- No vision inference support
- Large context length may exceed typical GPU limits
Best for
- Enterprise language understanding
- Research on MoE architectures
- Long‑form content creation
Who it’s for. Developers and researchers needing a high‑capacity MoE language model for text tasks. The model is suitable for environments with sufficient GPU memory and a need for advanced reasoning capabilities.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.3 GB | 2.6 GB | Fits | Get file → | |
| IQ1_M | 67.1 GB (5 parts) | 69.4 GB | Too big | Get file → | |
| IQ2_XXS | 78.3 GB (6 parts) | 80.6 GB | Too big | Get file → | |
| IQ2_XS | 87.4 GB (7 parts) | 89.7 GB | Too big | Get file → | |
| IQ2_S | 88.7 GB (7 parts) | 91.0 GB | Too big | Get file → | |
| Q2_K | 109.1 GB (8 parts) | 111.4 GB | Too big | Get file → | |
| Q3_K_M | 142.6 GB (11 parts) | 144.9 GB | Too big | Get file → | |
| Q4_K_M | 180.2 GB (14 parts) | 182.5 GB | Too big | Get file → | |
| Q5_K_M | 212.0 GB (16 parts) | 214.3 GB | Too big | Get file → | |
| Q6_K | 245.8 GB (20 parts) | 248.1 GB | Too big | Get file → | |
| Q8_0 | 318.3 GB (26 parts) | 320.6 GB | Too big | Get file → | |
| F16 | 599.0 GB (39 parts) | 601.3 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_M (67.1 GB file, ~69 GB memory at 8 k context). For 16 GB VRAM, Q4_K_M (180.16 GB file, ~182 GB memory) is a balanced choice. For 32 GB VRAM, Q6_K (245.76 GB file, ~248 GB memory) provides higher precision. Memory estimates are approximate.
What GPU do you need for baidu.ERNIE 4.5 300B A47B PT?
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: the best quant that fits is the main file (0.3 GB file, about 2.6 GB in use).
- 6 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 2.6 GB in use).
- 8 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 2.6 GB in use).
- 12 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 2.6 GB in use).
- 16 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 2.6 GB in use).
- 24 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 2.6 GB in use).
- 32 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 2.6 GB in use).
- 48 GB VRAM: the best quant that fits is the main file (0.3 GB file, about 2.6 GB in use).
Which baidu.ERNIE 4.5 300B A47B PT quant should you choose?
Q4_K_M is the usual starting point for baidu.ERNIE 4.5 300B A47B PT: it balances file size and quality well. Here it is a 180.2 GB file.
The Q8_0 file is 318.3 GB, about 1.8 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is the first one in the table at 0.3 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about baidu.ERNIE 4.5 300B A47B PT
How much VRAM does baidu.ERNIE 4.5 300B A47B PT need?
About 182.5 GB at Q4_K_M with an 8K context window. That is the 180.2 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 baidu.ERNIE 4.5 300B A47B PT on an 8 GB GPU?
Yes, with the smaller files: the 0.3 GB file needs about 2.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of baidu.ERNIE 4.5 300B A47B PT GGUF?
It depends on the quant: file is 0.3 GB, IQ1_M is 67.1 GB, IQ2_XXS is 78.3 GB, IQ2_XS is 87.4 GB, IQ2_S is 88.7 GB, Q2_K is 109.1 GB and 6 more.
What context length does baidu.ERNIE 4.5 300B A47B PT support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.8 GB at 8K and 7.2 GB at 32K.
Where can I download baidu.ERNIE 4.5 300B A47B PT GGUF?
The files are on Hugging Face in the DevQuasar/baidu.ERNIE-4.5-300B-A47B-PT-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 baidu.ERNIE 4.5 300B A47B PT 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.
Can I run this model on 8 GB VRAM?
No, the smallest quantized file requires 67.1 GB and ~69 GB memory at 8 k context. 8 GB VRAM is insufficient.
Does this model support vision tasks?
No, the provided model is text‑only; vision support is not included.
What license applies to this model?
The model is released under the Apache‑2.0 license.