Hari Q2.5 needs about 50.6 GB of memory at Q4_K_M (a 47.4 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 7 quants, from 29.8 GB (Q2_K) to 47.4 GB (Q4_K_M). Longer context needs more: at 32K tokens the KV cache alone adds about 10.7 GB.
Hari Q2.5 is a 72.7B-parameter Hari model in GGUF format, quantized by mradermacher. Its trained context length is 128K tokens. The repository has 7 quant files, from 29.8 GB (Q2_K) to 47.4 GB (Q4_K_M).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 29.8 GB | 33.0 GB | Too big | Get file → | |
| Q3_K_S | 34.5 GB | 37.7 GB | Too big | Get file → | |
| Q3_K_M | 37.7 GB | 40.9 GB | Too big | Get file → | |
| Q3_K_L | 39.5 GB | 42.7 GB | Too big | Get file → | |
| IQ4_XS | 40.2 GB | 43.3 GB | Too big | Get file → | |
| Q4_K_S | 43.9 GB | 47.1 GB | Too big | Get file → | |
| Q4_K_M | 47.4 GB | 50.6 GB | Too big | Get file → |
What GPU do you need for Hari Q2.5?
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_S (43.9 GB file, about 47.1 GB in use).
Which Hari Q2.5 quant should you choose?
Q4_K_M is the usual starting point for Hari Q2.5: it balances file size and quality well. Here it is a 47.4 GB file.
If memory is tight, the smallest file is Q2_K at 29.8 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 Hari Q2.5
How much VRAM does Hari Q2.5 need?
About 50.6 GB at Q4_K_M with an 8K context window. That is the 47.4 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 Hari Q2.5 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 33.0 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Hari Q2.5 GGUF?
It depends on the quant: Q2_K is 29.8 GB, Q3_K_S is 34.5 GB, Q3_K_M is 37.7 GB, Q3_K_L is 39.5 GB, IQ4_XS is 40.2 GB, Q4_K_S is 43.9 GB and 1 more.
What context length does Hari Q2.5 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 Hari Q2.5 GGUF?
The files are on Hugging Face in the mradermacher/hari-q2.5-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 Hari Q2.5 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.