ERNIE 4.5 37B A3B Thinking Brainstorm20x needs about 23.8 GB of memory at Q4_K_M (a 22.5 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 11 quants, from 13.6 GB (Q2_K) to 39.5 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 3.2 GB.
ERNIE 4.5 37B A3B Thinking Brainstorm20x is a 37.2B-parameter Ernie model in GGUF format, quantized by mradermacher. Its trained context length is 128K tokens. The repository has 11 quant files, from 13.6 GB (Q2_K) to 39.5 GB (Q8_0).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 13.6 GB | 14.9 GB | Too big | Get file → | |
| Q3_K_S | 16.1 GB | 17.4 GB | Too big | Get file → | |
| Q3_K_M | 17.8 GB | 19.1 GB | Too big | Get file → | |
| Q3_K_L | 19.3 GB | 20.6 GB | Too big | Get file → | |
| IQ4_XS | 20.0 GB | 21.3 GB | Too big | Get file → | |
| Q4_K_S | 21.1 GB | 22.4 GB | Too big | Get file → | |
| Q4_K_M | 22.5 GB | 23.8 GB | Too big | Get file → | |
| Q5_K_S | 25.6 GB | 26.9 GB | Too big | Get file → | |
| Q5_K_M | 26.4 GB | 27.7 GB | Too big | Get file → | |
| Q6_K | 30.5 GB | 31.8 GB | Too big | Get file → | |
| Q8_0 | 39.5 GB | 40.8 GB | Too big | Get file → |
What GPU do you need for ERNIE 4.5 37B A3B Thinking Brainstorm20x?
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 Q2_K (13.6 GB file, about 14.9 GB in use).
- 24 GB VRAM: the best quant that fits is Q4_K_M (22.5 GB file, about 23.8 GB in use).
- 32 GB VRAM: the best quant that fits is Q6_K (30.5 GB file, about 31.8 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (39.5 GB file, about 40.8 GB in use).
Which ERNIE 4.5 37B A3B Thinking Brainstorm20x quant should you choose?
Q4_K_M is the usual starting point for ERNIE 4.5 37B A3B Thinking Brainstorm20x: it balances file size and quality well. Here it is a 22.5 GB file.
The Q8_0 file is 39.5 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 Q2_K at 13.6 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Frequently asked questions about ERNIE 4.5 37B A3B Thinking Brainstorm20x
How much VRAM does ERNIE 4.5 37B A3B Thinking Brainstorm20x need?
About 23.8 GB at Q4_K_M with an 8K context window. That is the 22.5 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 37B A3B Thinking Brainstorm20x on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 14.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of ERNIE 4.5 37B A3B Thinking Brainstorm20x GGUF?
It depends on the quant: Q2_K is 13.6 GB, Q3_K_S is 16.1 GB, Q3_K_M is 17.8 GB, Q3_K_L is 19.3 GB, IQ4_XS is 20.0 GB, Q4_K_S is 21.1 GB and 5 more.
What context length does ERNIE 4.5 37B A3B Thinking Brainstorm20x support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.8 GB at 8K and 3.2 GB at 32K.
Where can I download ERNIE 4.5 37B A3B Thinking Brainstorm20x GGUF?
The files are on Hugging Face in the mradermacher/ERNIE-4.5-37B-A3B-Thinking-Brainstorm20x-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 37B A3B Thinking Brainstorm20x 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.