Phi 3 medium 128k instruct IMat needs about 8.7 GB of memory at IQ3_M (a 6.5 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 25 quants, from 3.0 GB (IQ1_S) to 27.9 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 6.7 GB.
Phi‑3 Medium 128k Instruct IMat – 14B parameter LLM
Phi‑3 Medium 128k Instruct IMat is a 14B parameter model with a 131072 token context window. It is quantized for llama.cpp using IMatrix and supports a range of quantization levels. The model is released under the MIT license.
Good at
- Reasoning and logic
- Code generation
- Long‑form text
Limits
- No vision support
- Requires 8k context for memory estimates
- License is MIT
Best for
- Chatbots with long context
- Research prototyping
- Low‑latency inference
Who it’s for. Developers and researchers needing a medium‑size Phi‑3 model for English‑only applications. The model is suitable for environments with limited memory and strict latency requirements.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 3.0 GB | 5.1 GB | Fits | Get file → | |
| IQ1_M | 3.2 GB | 5.4 GB | Fits | Get file → | |
| IQ2_XXS | 3.7 GB | 5.9 GB | Fits | Get file → | |
| IQ2_XS | 4.1 GB | 6.3 GB | Fits | Get file → | |
| IQ2_S | 4.3 GB | 6.5 GB | Fits | Get file → | |
| IQ2_M | 4.7 GB | 6.9 GB | Fits | Get file → | |
| Q2_K_S | 4.8 GB | 6.9 GB | Fits | Get file → | |
| Q2_K | 5.1 GB | 7.3 GB | Tight | Get file → | |
| IQ3_XXS | 5.5 GB | 7.6 GB | Tight | Get file → | |
| IQ3_XS | 5.8 GB | 8.0 GB | Tight | Get file → | |
| IQ3_S | 6.1 GB | 8.2 GB | Too big | Get file → | |
| Q3_K_S | 6.1 GB | 8.2 GB | Too big | Get file → | |
| IQ3_M | 6.5 GB | 8.7 GB | Too big | Get file → | |
| Q3_K | 6.9 GB | 9.1 GB | Too big | Get file → | |
| IQ4_XS | 7.5 GB | 9.6 GB | Too big | Get file → | |
| Q3_K_L | 7.5 GB | 9.7 GB | Too big | Get file → | |
| IQ4_NL | 7.9 GB | 10.1 GB | Too big | Get file → | |
| Q4_K_S | 8.0 GB | 10.1 GB | Too big | Get file → | |
| Q4_K | 8.6 GB | 10.7 GB | Too big | Get file → | |
| Q5_K_S | 9.6 GB | 11.8 GB | Too big | Get file → | |
| Q5_K | 10.1 GB | 12.3 GB | Too big | Get file → | |
| Q6_K | 11.5 GB | 13.6 GB | Too big | Get file → | |
| Q8_0 | 14.8 GB | 17.0 GB | Too big | Get file → | |
| BF16 | 27.9 GB | 30.1 GB | Too big | Get file → | |
| — | 27.9 GB | 30.1 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_S (2.96 GB) or IQ1_M (3.24 GB). For 12 GB VRAM, IQ2_M (4.72 GB) or Q2_K (5.14 GB) are good choices. For 16 GB VRAM, Q3_K (6.92 GB) or Q4_K (8.57 GB) provide higher quality.
What GPU do you need for Phi 3 medium 128k instruct IMat?
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 IQ2_XXS (3.7 GB file, about 5.9 GB in use).
- 8 GB VRAM: the best quant that fits is IQ3_XS (5.8 GB file, about 8.0 GB in use).
- 12 GB VRAM: the best quant that fits is Q5_K_S (9.6 GB file, about 11.8 GB in use).
- 16 GB VRAM: the best quant that fits is Q6_K (11.5 GB file, about 13.6 GB in use).
- 24 GB VRAM: the best quant that fits is Q8_0 (14.8 GB file, about 17.0 GB in use).
- 32 GB VRAM: the best quant that fits is BF16 (27.9 GB file, about 30.1 GB in use).
Which Phi 3 medium 128k instruct IMat quant should you choose?
IQ3_M (6.5 GB) is the middle option in this repository. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.
If memory is tight, the smallest file is IQ1_S at 3.0 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 Phi 3 medium 128k instruct IMat
How much VRAM does Phi 3 medium 128k instruct IMat need?
About 8.7 GB at IQ3_M with an 8K context window. That is the 6.5 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 Phi 3 medium 128k instruct IMat on an 8 GB GPU?
Yes, with IQ3_XS: the 5.8 GB file needs about 8.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Phi 3 medium 128k instruct IMat GGUF?
It depends on the quant: IQ1_S is 3.0 GB, IQ1_M is 3.2 GB, IQ2_XXS is 3.7 GB, IQ2_XS is 4.1 GB, IQ2_S is 4.3 GB, IQ2_M is 4.7 GB and 19 more.
What context length does Phi 3 medium 128k instruct IMat support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.7 GB at 8K and 6.7 GB at 32K.
Where can I download Phi 3 medium 128k instruct IMat GGUF?
The files are on Hugging Face in the legraphista/Phi-3-medium-128k-instruct-IMat-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 Phi 3 medium 128k instruct IMat 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.
Can I run this model on 8 GB VRAM?
Yes, using IQ1_S or IQ1_M quantizations will fit within 8 GB VRAM with approximate 5.1–5.4 GB memory usage at 8k context.
Does this model support vision tasks?
No, the model does not include vision capabilities.
What is the maximum context length?
The model supports up to 131072 tokens as its trained context length.