PH

Phi4 Abliterated

PhiGGUF by mradermacherUpdated 2025-05-11
Download on Hugging Face
Parameters14.7B
Context16K
Layers40
Architecturephi3
Smallest quant5.5 GB
Quick answer

Phi4 Abliterated needs about 11.2 GB of memory at Q4_K_M (a 9.1 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 22 quants, from 5.5 GB (Q2_K) to 15.6 GB (Q8_0).

Phi4 Abliterated – a 14.7B Phi3 model with neutral fine‑tuning

Phi4 Abliterated is a 14.7B parameter Phi3 model trained with a new layer‑wise refusal direction approach. It supports a 16384 token context window and is designed for neutral response generation. The model is available in multiple GGUF quantizations.

Good at

  • Neutral response generation
  • Fine‑tuning friendly
  • Large context support (16384 tokens)

Limits

  • No vision support
  • License not specified
  • Requires 8k context memory estimates

Best for

  • Fine‑tuning for reduced censorship
  • Chat applications needing neutrality
  • Research on refusal direction methods

Who it’s for. Developers building neutral LLMs and researchers exploring layer‑wise refusal mechanisms. The model is suitable for projects that need a foundation for fine‑tuning to balance usability and neutrality.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K5.5 GB 7.7 GB
Tight Get file →
Q2_K5.5 GB 7.7 GB
Tight Get file →
Q3_K_S6.5 GB 8.7 GB
Too big Get file →
Q3_K_S6.5 GB 8.7 GB
Too big Get file →
Q3_K_M7.4 GB 9.5 GB
Too big Get file →
Q3_K_M7.4 GB 9.5 GB
Too big Get file →
Q3_K_L7.9 GB 10.1 GB
Too big Get file →
Q3_K_L7.9 GB 10.1 GB
Too big Get file →
IQ4_XS8.0 GB 10.2 GB
Too big Get file →
IQ4_XS8.0 GB 10.2 GB
Too big Get file →
Q4_K_S8.4 GB 10.6 GB
Too big Get file →
Q4_K_S8.4 GB 10.6 GB
Too big Get file →
Q4_K_M9.1 GB 11.2 GB
Too big Get file →
Q4_K_M9.1 GB 11.2 GB
Too big Get file →
Q5_K_S10.2 GB 12.3 GB
Too big Get file →
Q5_K_S10.2 GB 12.3 GB
Too big Get file →
Q5_K_M10.6 GB 12.8 GB
Too big Get file →
Q5_K_M10.6 GB 12.8 GB
Too big Get file →
Q6_K12.0 GB 14.2 GB
Too big Get file →
Q6_K12.0 GB 14.2 GB
Too big Get file →
Q8_015.6 GB 17.8 GB
Too big Get file →
Q8_015.6 GB 17.8 GB
Too big Get file →

For 8GB VRAM, use Q2_K (~5.55GB) which estimates 7.7GB memory at 8k context. If more memory is available, Q3_K_S (~6.5GB) uses 8.7GB, and Q4_K_M (~9.05GB) uses 11.2GB. Choose the smallest quant that fits your VRAM budget.

What GPU do you need for Phi4 Abliterated?

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: the best quant that fits is Q2_K (5.5 GB file, about 7.7 GB in use).
  • 12 GB VRAM: the best quant that fits is Q4_K_M (9.1 GB file, about 11.2 GB in use).
  • 16 GB VRAM: the best quant that fits is Q6_K (12.0 GB file, about 14.2 GB in use).
  • 24 GB VRAM: the best quant that fits is Q8_0 (15.6 GB file, about 17.8 GB in use).

Which Phi4 Abliterated quant should you choose?

Q4_K_M is the usual starting point for Phi4 Abliterated: it balances file size and quality well. Here it is a 9.1 GB file.

The Q8_0 file is 15.6 GB, about 1.7 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 Q2_K at 5.5 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

Architecturephi3
Parameters14.7B
Layers40
Trained context16,384 tokens
KV cache (fp16)200 KB per token

Source

Quantized bymradermacher
FamilyPhi
Downloads (HF)678
Last updated2025-05-11

Frequently asked questions about Phi4 Abliterated

How much VRAM does Phi4 Abliterated need?

About 11.2 GB at Q4_K_M with an 8K context window. That is the 9.1 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 Phi4 Abliterated on an 8 GB GPU?

Yes, with Q2_K: the 5.5 GB file needs about 7.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Phi4 Abliterated GGUF?

It depends on the quant: Q2_K is 5.5 GB, Q2_K is 5.5 GB, Q3_K_S is 6.5 GB, Q3_K_S is 6.5 GB, Q3_K_M is 7.4 GB, Q3_K_M is 7.4 GB and 16 more.

What context length does Phi4 Abliterated support?

The GGUF header lists a trained context length of 16,384 tokens (16K). The KV cache grows with the context you use: about 1.7 GB at 8K and 3.4 GB at 16K.

Where can I download Phi4 Abliterated GGUF?

The files are on Hugging Face in the mradermacher/Phi4-abliterated-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Can I run Phi4 Abliterated on 8GB VRAM?

Yes, the Q2_K quant (~5.55GB) uses about 7.7GB memory at 8k context, fitting within 8GB VRAM.

Does Phi4 Abliterated support vision?

No, the model has no vision capabilities.

What is the maximum context length?

The model supports up to 16384 tokens of context.

Other GGUFs of Phi4-abliterated

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Phi models

All Phi →

Other medium models

All medium models →