Downloads · 30 days
236
27% of all-time downloads
liodon-ai/qwen3-4b-heretic-imatrix-GGUF
qwen3-4b-heretic-imatrix-GGUF is a text generation model from liodon-ai. Use it when you need the model to write or continue text. The card lists the license as other.
GGUF quantizations of DreamFast/qwen3-4b-heretic, published by Liodon AI.
Downloads · 30 days
236
27% of all-time downloads
All-time downloads
877
Public
Repo size
54.4 GB
Likes
0
Public
Click a slice to open those files.
.gguf54.4 GB · 100%
From the Hugging Face model README
GGUF quantizations of DreamFast/qwen3-4b-heretic, published by Liodon AI.
llama.cpp
llama-cli -hf liodon-ai/qwen3-4b-heretic-imatrix-GGUF:Q4_K_M
Ollama
ollama run hf.co/liodon-ai/qwen3-4b-heretic-imatrix-GGUF:Q4_K_M
LM Studio / Jan — search liodon-ai/qwen3-4b-heretic-imatrix-GGUF and pick your quant.
| Quant | Size | VRAM est. | Notes |
|---|---|---|---|
IQ2_M | 4.31 GB | ~5 GB | 2-bit, iMatrix — smallest usable |
IQ3_M | 5.66 GB | ~7 GB | 3-bit, iMatrix — great quality/size tradeoff |
IQ4_XS | 6.55 GB | ~8 GB | 4-bit extra-small, iMatrix |
Q4_K_M | 7.30 GB | ~8 GB | 4-bit, iMatrix-calibrated (recommended) |
Q5_K_M | 8.45 GB | ~10 GB | 5-bit, iMatrix-calibrated |
Q6_K | 9.66 GB | ~11 GB | 6-bit, iMatrix-calibrated, near-lossless |
Q8_0 | 12.51 GB | ~14 GB | 8-bit, essentially lossless |
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/qwen3-4b-heretic-GGUF
Quantized by Liodon AI