CobrIX 1.5 preview Coder Full 72B A18B needs about 45.2 GB of memory at Q4_K_M (a 43.7 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 2 quants, from 43.7 GB (Q4_K_M) to 51.1 GB (Q5_K_M). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
CobrIX 1.5 preview Coder Full 72B A18B is a 71.8B-parameter Cobrix model in GGUF format, quantized by CobrIX. Its trained context length is 1024K tokens. The repository has 2 quant files, from 43.7 GB (Q4_K_M) to 51.1 GB (Q5_K_M).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 43.7 GB | 45.2 GB | Too big | Get file → | |
| Q5_K_M | 51.1 GB | 52.6 GB | Too big | Get file → |
What GPU do you need for CobrIX 1.5 preview Coder Full 72B A18B?
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: no quant fits at an 8K context.
- 24 GB VRAM: no quant fits at an 8K context.
- 32 GB VRAM: no quant fits at an 8K context.
- 48 GB VRAM: the best quant that fits is Q4_K_M (43.7 GB file, about 45.2 GB in use).
Which CobrIX 1.5 preview Coder Full 72B A18B quant should you choose?
Q4_K_M is the usual starting point for CobrIX 1.5 preview Coder Full 72B A18B: it balances file size and quality well. Here it is a 43.7 GB file.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Frequently asked questions about CobrIX 1.5 preview Coder Full 72B A18B
How much VRAM does CobrIX 1.5 preview Coder Full 72B A18B need?
About 45.2 GB at Q4_K_M with an 8K context window. That is the 43.7 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 CobrIX 1.5 preview Coder Full 72B A18B on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 45.2 GB, so you would need a larger GPU or CPU offloading.
What is the file size of CobrIX 1.5 preview Coder Full 72B A18B GGUF?
It depends on the quant: Q4_K_M is 43.7 GB, Q5_K_M is 51.1 GB.
What context length does CobrIX 1.5 preview Coder Full 72B A18B support?
The GGUF header lists a trained context length of 1,048,576 tokens (1024K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.
Where can I download CobrIX 1.5 preview Coder Full 72B A18B GGUF?
The files are on Hugging Face in the CobrIX/CobrIX-1.5-preview-Coder-Full-72B-A18B-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 CobrIX 1.5 preview Coder Full 72B A18B use?
Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.