Kimidev needs about 60.0 GB of memory at TQ2_0 (a 56.9 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 3 quants, from 54.5 GB (TQ1_0) to 77.3 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 10.7 GB.
Kimidev is a 72.7B-parameter Kimi model in GGUF format, quantized by volker-mauel. Its trained context length is 128K tokens. The repository has 3 quant files, from 54.5 GB (TQ1_0) to 77.3 GB (Q8_0).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| TQ1_0 | 54.5 GB (3 parts) | 57.7 GB | Too big | Get file → | |
| TQ2_0 | 56.9 GB (3 parts) | 60.0 GB | Too big | Get file → | |
| Q8_0 | 77.3 GB (4 parts) | 80.4 GB | Too big | Get file → |
What GPU do you need for Kimidev?
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: no quant fits at an 8K context.
Which Kimidev quant should you choose?
TQ2_0 (56.9 GB) is the middle option in this repository. Lower bit counts mean smaller files and more quality loss.
If memory is tight, the smallest file is TQ1_0 at 54.5 GB, with the quality trade-offs that come with fewer bits. Lower bit counts mean smaller files and more quality loss.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Kimidev
How much VRAM does Kimidev need?
About 60.0 GB at TQ2_0 with an 8K context window. That is the 56.9 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 Kimidev on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 57.7 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Kimidev GGUF?
It depends on the quant: TQ1_0 is 54.5 GB, TQ2_0 is 56.9 GB, Q8_0 is 77.3 GB.
What context length does Kimidev support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 2.7 GB at 8K and 10.7 GB at 32K.
Where can I download Kimidev GGUF?
The files are on Hugging Face in the volker-mauel/Kimi-Dev-72B-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 Kimidev 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.