phi3 lora sft mathinstruct lima i1 needs about 6.1 GB of memory at Q4_K_M (a 2.4 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 24 quants, from 0.8 GB (IQ1_S) to 3.1 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 12.9 GB.
phi3 lora sft mathinstruct lima i1 is a 3.82B-parameter Phi model in GGUF format, quantized by mradermacher. Its trained context length is 128K tokens. The repository has 24 quant files, from 0.8 GB (IQ1_S) to 3.1 GB (Q6_K).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 0.8 GB | 4.6 GB | Fits | Get file → | |
| IQ1_M | 0.9 GB | 4.6 GB | Fits | Get file → | |
| IQ2_XXS | 1.0 GB | 4.8 GB | Fits | Get file → | |
| IQ2_XS | 1.2 GB | 4.9 GB | Fits | Get file → | |
| IQ2_S | 1.2 GB | 4.9 GB | Fits | Get file → | |
| IQ2_M | 1.3 GB | 5.0 GB | Fits | Get file → | |
| Q2_K_S | 1.3 GB | 5.0 GB | Fits | Get file → | |
| Q2_K | 1.4 GB | 5.1 GB | Fits | Get file → | |
| IQ3_XXS | 1.5 GB | 5.2 GB | Fits | Get file → | |
| IQ3_XS | 1.6 GB | 5.3 GB | Fits | Get file → | |
| IQ3_S | 1.7 GB | 5.4 GB | Fits | Get file → | |
| Q3_K_S | 1.7 GB | 5.4 GB | Fits | Get file → | |
| IQ3_M | 1.9 GB | 5.6 GB | Fits | Get file → | |
| Q3_K_M | 2.0 GB | 5.7 GB | Fits | Get file → | |
| IQ4_XS | 2.1 GB | 5.8 GB | Fits | Get file → | |
| Q3_K_L | 2.1 GB | 5.8 GB | Fits | Get file → | |
| IQ4_NL | 2.2 GB | 5.9 GB | Fits | Get file → | |
| Q4_0 | 2.2 GB | 5.9 GB | Fits | Get file → | |
| Q4_K_S | 2.2 GB | 5.9 GB | Fits | Get file → | |
| Q4_K_M | 2.4 GB | 6.1 GB | Fits | Get file → | |
| Q4_1 | 2.4 GB | 6.1 GB | Fits | Get file → | |
| Q5_K_S | 2.6 GB | 6.4 GB | Fits | Get file → | |
| Q5_K_M | 2.8 GB | 6.5 GB | Fits | Get file → | |
| Q6_K | 3.1 GB | 6.9 GB | Fits | Get file → |
What GPU do you need for phi3 lora sft mathinstruct lima i1?
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: the best quant that fits is Q4_K_S (2.2 GB file, about 5.9 GB in use).
- 8 GB VRAM: the best quant that fits is Q6_K (3.1 GB file, about 6.9 GB in use).
Which phi3 lora sft mathinstruct lima i1 quant should you choose?
Q4_K_M is the usual starting point for phi3 lora sft mathinstruct lima i1: it balances file size and quality well. Here it is a 2.4 GB file.
If memory is tight, the smallest file is IQ1_S at 0.8 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 phi3 lora sft mathinstruct lima i1
How much VRAM does phi3 lora sft mathinstruct lima i1 need?
About 6.1 GB at Q4_K_M with an 8K context window. That is the 2.4 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 phi3 lora sft mathinstruct lima i1 on an 8 GB GPU?
Yes, with Q6_K: the 3.1 GB file needs about 6.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of phi3 lora sft mathinstruct lima i1 GGUF?
It depends on the quant: IQ1_S is 0.8 GB, IQ1_M is 0.9 GB, IQ2_XXS is 1.0 GB, IQ2_XS is 1.2 GB, IQ2_S is 1.2 GB, IQ2_M is 1.3 GB and 18 more.
What context length does phi3 lora sft mathinstruct lima i1 support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 3.2 GB at 8K and 12.9 GB at 32K.
Where can I download phi3 lora sft mathinstruct lima i1 GGUF?
The files are on Hugging Face in the mradermacher/phi3_lora_sft-mathinstruct-lima-i1-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.