whittle dev needs about 39.0 GB of memory at Q8_0 (a 37.8 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 5 quants, from 37.8 GB (Q8_0) to 37.8 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 2.7 GB.
whittle dev is a 35.5B-parameter Whittle model in GGUF format, quantized by logic65. Its trained context length is 256K tokens. The repository has 5 quant files, from 37.8 GB (Q8_0) to 37.8 GB (Q8_0).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q8_0 | 37.8 GB | 39.0 GB | Too big | Get file → | |
| Q8_0 | 37.8 GB | 39.0 GB | Too big | Get file → | |
| Q8_0 | 37.8 GB | 39.0 GB | Too big | Get file → | |
| Q8_0 | 37.8 GB | 39.0 GB | Too big | Get file → | |
| Q8_0 | 37.8 GB | 39.0 GB | Too big | Get file → |
What GPU do you need for whittle dev?
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 Q8_0 (37.8 GB file, about 39.0 GB in use).
Which whittle dev quant should you choose?
Q8_0 (37.8 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.
If memory is tight, the smallest file is Q8_0 at 37.8 GB, with the quality trade-offs that come with fewer bits. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about whittle dev
How much VRAM does whittle dev need?
About 39.0 GB at Q8_0 with an 8K context window. That is the 37.8 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 whittle dev on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 39.0 GB, so you would need a larger GPU or CPU offloading.
What is the file size of whittle dev GGUF?
It depends on the quant: Q8_0 is 37.8 GB, Q8_0 is 37.8 GB, Q8_0 is 37.8 GB, Q8_0 is 37.8 GB, Q8_0 is 37.8 GB.
What context length does whittle dev 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 whittle dev GGUF?
The files are on Hugging Face in the logic65/whittle-dev repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.