---

license: mit

base_model: microsoft/phi-4

base_model_relation: quantized

library_name: gguf

pipeline_tag: text-generation

tags:

• gguf
• llama.cpp
• dashq
• quantized

---

!DASH-Q

phi-4 - DASH-Q 2-bit GGUF

A llama.cpp GGUF of microsoft/phi-4 quantized with

DASH-Q. It is a standard Q2_K-class mixed-precision file

(Q2_K / Q3_K / Q4_K tensors, about 3 bits per weight) and loads in any recent llama.cpp build.

| File | Size |

| --- | --- |

| phi-4-DASHQ-Q2_K_XL.gguf | 6.06 GB |

Perplexity (lower is better)

| Model | Size | WikiText-2 | C4 |

| --- | --- | --- | --- |

| F16 | 29.3 GB | 5.78 | 11.27 |

| llama.cpp Q2_K (imatrix) | 5.92 GB | 6.42 | 12.17 |

| unsloth Q2_K_L | 5.73 GB | 6.74 | 12.61 |

| DASH-Q (this file) | 6.06 GB | 6.18 | 11.87 |

llama-perplexity, context 2048; WikiText-2 test, C4 validation (256 x 2048 tokens).

Zero-shot accuracy (lm-eval-harness, 0-shot)

| arc_easy | arc_challenge | piqa | hellaswag | winogrande | openbookqa | commonsense_qa | truthfulqa_mc2 | lambada_openai | avg |

| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |

| 74.4 | 55.5 | 81.4 | 80.2 | 76.9 | 44.6 | 76.2 | 56.4 | 74.2 | 68.86 |

Usage

llama-cli -m phi-4-DASHQ-Q2_K_XL.gguf -ngl 99 -c 8192

License

Inherits the license of the base model (microsoft/phi-4).