ultiima 52B needs about 35.6 GB of memory at Q4_K_M (a 31.6 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 10 quants, from 19.4 GB (Q2_K) to 42.9 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 14.0 GB.
ultiima 52B is a 52.3B-parameter Ultiima model in GGUF format, quantized by mradermacher. Its trained context length is 128K tokens. The repository has 10 quant files, from 19.4 GB (Q2_K) to 42.9 GB (Q6_K).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 19.4 GB | 23.4 GB | Too big | Get file → | |
| Q3_K_S | 22.8 GB | 26.8 GB | Too big | Get file → | |
| Q3_K_M | 25.3 GB | 29.3 GB | Too big | Get file → | |
| Q3_K_L | 27.4 GB | 31.4 GB | Too big | Get file → | |
| IQ4_XS | 28.4 GB | 32.4 GB | Too big | Get file → | |
| Q4_K_S | 29.8 GB | 33.8 GB | Too big | Get file → | |
| Q4_K_M | 31.6 GB | 35.6 GB | Too big | Get file → | |
| Q5_K_S | 36.0 GB | 40.0 GB | Too big | Get file → | |
| Q5_K_M | 37.1 GB | 41.1 GB | Too big | Get file → | |
| Q6_K | 42.9 GB | 46.9 GB | Too big | Get file → |
What GPU do you need for ultiima 52B?
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: the best quant that fits is Q2_K (19.4 GB file, about 23.4 GB in use).
- 32 GB VRAM: the best quant that fits is Q3_K_L (27.4 GB file, about 31.4 GB in use).
- 48 GB VRAM: the best quant that fits is Q6_K (42.9 GB file, about 46.9 GB in use).
Which ultiima 52B quant should you choose?
Q4_K_M is the usual starting point for ultiima 52B: it balances file size and quality well. Here it is a 31.6 GB file.
If memory is tight, the smallest file is Q2_K at 19.4 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
Source
Frequently asked questions about ultiima 52B
How much VRAM does ultiima 52B need?
About 35.6 GB at Q4_K_M with an 8K context window. That is the 31.6 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 ultiima 52B on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 23.4 GB, so you would need a larger GPU or CPU offloading.
What is the file size of ultiima 52B GGUF?
It depends on the quant: Q2_K is 19.4 GB, Q3_K_S is 22.8 GB, Q3_K_M is 25.3 GB, Q3_K_L is 27.4 GB, IQ4_XS is 28.4 GB, Q4_K_S is 29.8 GB and 4 more.
What context length does ultiima 52B support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 3.5 GB at 8K and 14.0 GB at 32K.
Where can I download ultiima 52B GGUF?
The files are on Hugging Face in the mradermacher/ultiima-52B-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.