Darwin 36B Opus needs about 17.5 GB of memory at its main file (a 16.4 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 7 quants, from 10.8 GB to 25.0 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 2.7 GB.
Darwin 36B Opus is a 34.7B-parameter Darwin model in GGUF format, quantized by el4. Its trained context length is 256K tokens. The repository has 7 quant files, from 10.8 GB to 25.0 GB.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 10.8 GB | 11.9 GB | Too big | Get file → | |
| — | 11.6 GB | 12.8 GB | Too big | Get file → | |
| — | 13.5 GB | 14.7 GB | Too big | Get file → | |
| — | 16.4 GB | 17.5 GB | Too big | Get file → | |
| — | 17.6 GB | 18.8 GB | Too big | Get file → | |
| — | 22.7 GB | 23.9 GB | Too big | Get file → | |
| — | 25.0 GB | 26.2 GB | Too big | Get file → |
What GPU do you need for Darwin 36B Opus?
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: the best quant that fits is the main file (10.8 GB file, about 11.9 GB in use).
- 16 GB VRAM: the best quant that fits is the main file (13.5 GB file, about 14.7 GB in use).
- 24 GB VRAM: the best quant that fits is the main file (22.7 GB file, about 23.9 GB in use).
- 32 GB VRAM: the best quant that fits is the main file (25.0 GB file, about 26.2 GB in use).
Which Darwin 36B Opus quant should you choose?
If memory is tight, the smallest file is the first one in the table at 10.8 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 Darwin 36B Opus
How much VRAM does Darwin 36B Opus need?
About 17.5 GB at its main file with an 8K context window. That is the 16.4 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 Darwin 36B Opus on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 11.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Darwin 36B Opus GGUF?
It depends on the quant: file is 10.8 GB, file is 11.6 GB, file is 13.5 GB, file is 16.4 GB, file is 17.6 GB, file is 22.7 GB and 1 more.
What context length does Darwin 36B Opus support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.7 GB at 8K and 2.7 GB at 32K.
Where can I download Darwin 36B Opus GGUF?
The files are on Hugging Face in the el4/Darwin-36B-Opus-OPAL-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 Darwin 36B Opus 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.