---

license: other

base_model: Yuhan123/vicuna-13b-self_consistency_random_var_5

base_model_relation: quantized

library_name: gguf

pipeline_tag: text-generation

quantized_by: liodon-ai

tags:

• gguf
• local-llm
• llama.cpp
• lm-studio
• quantized
• ollama
• imatrix
• sub-4-bit
• llama

---

vicuna-13b-self_consistency_random_var_5 — iMatrix GGUF

GGUF quantizations of Yuhan123/vicuna-13b-self_consistency_random_var_5, published by Liodon AI.

Quick Start

llama.cpp

llama-cli -hf liodon-ai/vicuna-13b-self_consistency_random_var_5-imatrix-GGUF:Q4_K_M

Ollama

ollama run hf.co/liodon-ai/vicuna-13b-self_consistency_random_var_5-imatrix-GGUF:Q4_K_M

LM Studio / Jan — search liodon-ai/vicuna-13b-self_consistency_random_var_5-imatrix-GGUF and pick your quant.

Quants

| Quant | Size | VRAM est. | Notes |

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

| IQ2_M | 4.52 GB | ~5 GB | 2-bit, iMatrix — smallest usable |

| IQ3_M | 5.98 GB | ~7 GB | 3-bit, iMatrix — great quality/size tradeoff |

| IQ4_XS | 6.96 GB | ~8 GB | 4-bit extra-small, iMatrix |

| Q4_K_M | 7.87 GB | ~9 GB | 4-bit, iMatrix-calibrated (recommended) |

| Q5_K_M | 9.23 GB | ~11 GB | 5-bit, iMatrix-calibrated |

| Q6_K | 10.68 GB | ~12 GB | 6-bit, iMatrix-calibrated, near-lossless |

| Q8_0 | 13.83 GB | ~16 GB | 8-bit, essentially lossless |

What is iMatrix?

Standard quantization treats all weights equally. iMatrix runs 128 calibration chunks through

the full-precision model to find which weights matter most, then allocates more precision where

it counts. At Q2/Q3/Q4 this means noticeably better coherence and instruction-following —

same file size, better output.

Calibration: 2M tokens of WikiText-103.

> Also see plain (non-iMatrix) quants: liodon-ai/vicuna-13b-self_consistency_random_var_5-GGUF

Source

• Model: Yuhan123/vicuna-13b-self_consistency_random_var_5
• License: other

Citation

@misc{liodonai_vicuna_13b_self_consistency_random_var_5_imatrix_gguf,
  title        = {vicuna-13b-self_consistency_random_var_5 — iMatrix GGUF},
  author       = {{Liodon AI}},
  year         = {2026},
  howpublished = {\url{https://huggingface.co/liodon-ai/vicuna-13b-self_consistency_random_var_5-imatrix-GGUF}},
  note         = {iMatrix GGUF quantization of Yuhan123/vicuna-13b-self_consistency_random_var_5}
}

---

Quantized by Liodon AI