Auto-Quantized GGUF Model

This repository contains automated GGUF quantization files for coder3101/gemma-4-E4B-it-heretic.

The calibration data for the imatrix is targeted at Chinese novels and role-playing (RP), while preserving logic and common sense.
imatrix ็š„ๆ กๅ‡†ๆ•ฐๆฎไปฅไธญๆ–‡็š„ๅฐ่ฏดใ€่ง’่‰ฒๆ‰ฎๆผ”ไธบ็›ฎๆ ‡๏ผŒๅŒๆ—ถไฟ็•™้€ป่พ‘ๅ’Œๅธธ่ฏ†ใ€‚

๐Ÿ“Š Perplexity Evaluation

(Tested against the provided calibration dataset)

  • Base (F16/BF16): PPL = 109.1430 +/- 1.51189
  • Q5_K_M: PPL = 445.1681 +/- 7.15013

This PPL measurement is definitely abnormal, and the cause is currently unknown.

Downloads last month
1,383
GGUF
Model size
8B params
Architecture
gemma4
Hardware compatibility
Log In to add your hardware

5-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for nuofang/gemma-4-E4B-it-heretic-GGUF

Quantized
(7)
this model