ERNIE 4.5 29B A4B PT needs about 18.7 GB of memory at Q4_K_M (a 17.5 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 24 quants, from 0.1 GB to 23.8 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 2.5 GB.
ERNIE 4.5 29B A4B PT – 29B parameter MoE model in GGUF format
ERNIE 4.5 29B A4B PT is a 29B parameter mixture‑of‑experts model with 4B active experts. It supports a 131072 token context window and is released under the apache‑2.0 license. The model is available in multiple GGUF quantizations.
Good at
- Large‑scale text generation
- Multimodal reasoning
- High‑capacity context handling
Limits
- Requires 8‑bit or higher quant for 8 GB VRAM
- No vision support in current GGUF
- License restricts commercial redistribution
Best for
- Research on long‑form content
- Enterprise chatbot back‑ends
- Academic language model studies
Who it’s for. Researchers and developers needing a large MoE language model with extensive context. The model is suitable for those who can manage the memory requirements and comply with the apache‑2.0 license.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.1 GB | 1.2 GB | Fits | Get file → | |
| IQ1_S | 5.9 GB | 7.0 GB | Fits | Get file → | |
| IQ1_M | 6.6 GB | 7.7 GB | Tight | Get file → | |
| IQ2_XXS | 7.7 GB | 8.8 GB | Too big | Get file → | |
| IQ2_XS | 8.5 GB | 9.7 GB | Too big | Get file → | |
| IQ2_S | 8.6 GB | 9.8 GB | Too big | Get file → | |
| IQ2_M | 9.5 GB | 10.6 GB | Too big | Get file → | |
| Q2_K_S | 9.9 GB | 11.0 GB | Too big | Get file → | |
| Q2_K | 10.7 GB | 11.8 GB | Too big | Get file → | |
| IQ3_XXS | 11.2 GB | 12.3 GB | Too big | Get file → | |
| IQ3_XS | 11.9 GB | 13.0 GB | Too big | Get file → | |
| Q3_K_S | 12.6 GB | 13.7 GB | Too big | Get file → | |
| IQ3_S | 12.6 GB | 13.7 GB | Too big | Get file → | |
| IQ3_M | 12.7 GB | 13.8 GB | Too big | Get file → | |
| Q3_K_M | 13.9 GB | 15.0 GB | Too big | Get file → | |
| Q3_K_L | 15.1 GB | 16.2 GB | Too big | Get file → | |
| IQ4_XS | 15.5 GB | 16.6 GB | Too big | Get file → | |
| Q4_0 | 16.4 GB | 17.6 GB | Too big | Get file → | |
| Q4_K_S | 16.5 GB | 17.6 GB | Too big | Get file → | |
| Q4_K_M | 17.5 GB | 18.7 GB | Too big | Get file → | |
| Q4_1 | 18.2 GB | 19.3 GB | Too big | Get file → | |
| Q5_K_S | 20.0 GB | 21.1 GB | Too big | Get file → | |
| Q5_K_M | 20.6 GB | 21.7 GB | Too big | Get file → | |
| Q6_K | 23.8 GB | 24.9 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_S (5.92 GB, 7 GB memory at 8 k context). For 16 GB VRAM, IQ2_XS (8.54 GB, 9.7 GB memory) or Q4_K_M (17.54 GB, 18.7 GB memory). For 32 GB VRAM, Q5_K_M (20.57 GB, 21.7 GB memory).
What GPU do you need for ERNIE 4.5 29B A4B 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 (0.1 GB file, about 1.2 GB in use).
- 6 GB VRAM: the best quant that fits is the main file (0.1 GB file, about 1.2 GB in use).
- 8 GB VRAM: the best quant that fits is IQ1_M (6.6 GB file, about 7.7 GB in use).
- 12 GB VRAM: the best quant that fits is Q2_K (10.7 GB file, about 11.8 GB in use).
- 16 GB VRAM: the best quant that fits is Q3_K_M (13.9 GB file, about 15.0 GB in use).
- 24 GB VRAM: the best quant that fits is Q5_K_M (20.6 GB file, about 21.7 GB in use).
- 32 GB VRAM: the best quant that fits is Q6_K (23.8 GB file, about 24.9 GB in use).
Which ERNIE 4.5 29B A4B PT quant should you choose?
Q4_K_M is the usual starting point for ERNIE 4.5 29B A4B PT: it balances file size and quality well. Here it is a 17.5 GB file.
If memory is tight, the smallest file is the first one in the table at 0.1 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 ERNIE 4.5 29B A4B PT
How much VRAM does ERNIE 4.5 29B A4B PT need?
About 18.7 GB at Q4_K_M with an 8K context window. That is the 17.5 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 29B A4B PT on an 8 GB GPU?
Yes, with IQ1_M: the 6.6 GB file needs about 7.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of ERNIE 4.5 29B A4B PT GGUF?
It depends on the quant: file is 0.1 GB, IQ1_S is 5.9 GB, IQ1_M is 6.6 GB, IQ2_XXS is 7.7 GB, IQ2_XS is 8.5 GB, IQ2_S is 8.6 GB and 18 more.
What context length does ERNIE 4.5 29B A4B 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 0.6 GB at 8K and 2.5 GB at 32K.
Where can I download ERNIE 4.5 29B A4B PT GGUF?
The files are on Hugging Face in the mradermacher/ERNIE-4.5-29B-A4B-PT-i1-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 29B A4B 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?
Use the IQ1_S quant, which is 5.92 GB and requires about 7 GB of VRAM at 8 k context. It fits within an 8 GB GPU with headroom for overhead.
Does the model support vision tasks?
The GGUF version does not include vision capabilities. It is a text‑only model.
What license applies to this model?
The model is released under the apache‑2.0 license, allowing free use with attribution.