Phi 4 mini instruct heretic needs about 4.1 GB of memory at Q4_K_M (a 2.5 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 7 quants, from 0.0 GB to 4.1 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
Phi 4 mini instruct heretic – a 3.8B Phi‑4‑mini model with removed refusal direction
Phi‑4‑mini‑instruct‑heretic is a 3.84B parameter Phi‑3 architecture model. It is an abliterated build of microsoft/Phi‑4‑mini‑instruct, retaining reasoning and math strengths while removing safety refusals. The model is MIT licensed and available in several GGUF quantizations.
Good at
- Reasoning and math tasks
- Instruction following
- Unrestricted local inference
Limits
- No vision support
- Requires 8‑k context for estimates
- MIT license limits commercial use without attribution
Best for
- Research on alignment techniques
- Creative writing without safety filters
- Local inference on modest hardware
Who it’s for. Researchers and developers needing a lightweight Phi‑4‑mini model that can comply with requests the original model would refuse. The model is suitable for local deployment and experimentation with alignment methods.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.0 GB | 1.6 GB | Fits | Get file → | |
| IQ3_M | 2.0 GB | 3.6 GB | Fits | Get file → | |
| IQ4_XS | 2.2 GB | 3.8 GB | Fits | Get file → | |
| Q4_K_M | 2.5 GB | 4.1 GB | Fits | Get file → | |
| Q5_K_M | 2.8 GB | 4.4 GB | Fits | Get file → | |
| Q6_K | 3.2 GB | 4.7 GB | Fits | Get file → | |
| Q8_0 | 4.1 GB | 5.7 GB | Fits | Get file → |
For 8‑GB VRAM, use IQ4_XS (2.22 GB) or Q4_K_M (2.49 GB). For 16‑GB VRAM, Q5_K_M (2.81 GB) or Q6_K (3.16 GB) provide a good balance of size and quality. Larger VRAM allows Q8_0 (4.08 GB) for the highest fidelity.
What GPU do you need for Phi 4 mini instruct heretic?
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: the best quant that fits is IQ4_XS (2.2 GB file, about 3.8 GB in use).
- 6 GB VRAM: the best quant that fits is Q8_0 (4.1 GB file, about 5.7 GB in use).
Which Phi 4 mini instruct heretic quant should you choose?
Q4_K_M is the usual starting point for Phi 4 mini instruct heretic: it balances file size and quality well. Here it is a 2.5 GB file.
The Q8_0 file is 4.1 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is the first one in the table at 0.0 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Phi 4 mini instruct heretic
How much VRAM does Phi 4 mini instruct heretic need?
About 4.1 GB at Q4_K_M with an 8K context window. That is the 2.5 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 4 mini instruct heretic on an 8 GB GPU?
Yes, with Q8_0: the 4.1 GB file needs about 5.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Phi 4 mini instruct heretic GGUF?
It depends on the quant: file is 0.0 GB, IQ3_M is 2.0 GB, IQ4_XS is 2.2 GB, Q4_K_M is 2.5 GB, Q5_K_M is 2.8 GB, Q6_K is 3.2 GB and 1 more.
What context length does Phi 4 mini instruct heretic support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.
Where can I download Phi 4 mini instruct heretic GGUF?
The files are on Hugging Face in the richardyoung/Phi-4-mini-instruct-heretic-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 4 mini instruct heretic 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, the IQ4_XS and Q4_K_M quantizations fit within 8 GB VRAM with approximate memory usage of 3.8 GB and 4.1 GB respectively.
Does this model support vision tasks?
No, the model has no vision capabilities and is text‑only.
What license applies to this model?
The model is MIT licensed, inherited from the base Microsoft Phi‑4‑mini model.