mannix/
qwq-32b-abilterated

1,378 9 months ago

QwQ is an experimental research model focused on advancing AI reasoning capabilities. Abliterated with uncensored prompt, i-matrix quants.

tools

Models

View all →

17 models

qwq-32b-abilterated:latest

19GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:Q3_K_S

14GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:Q3_K_L

17GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:Q4_0

19GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:q4_1

21GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:Q4_K_S

19GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:Q5_0

23GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:Q5_1

25GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:Q5_K_S

23GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:iq2_xxs

9.0GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:IQ2_XS

10.0GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:iq2_s

10GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:IQ3_XXS

13GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:IQ3_XS

14GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:IQ3_S

14GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:IQ4_XS

18GB · 32K context window · Text · 9 months ago

qwq-32b-abilterated:IQ4_NL

19GB · 32K context window · Text · 9 months ago

Readme

  • Quantization from fp32
  • Using i-matrix calibration_datav3.txt
  • Uncensored system prompt

This is an uncensored version of Qwen/QwQ-32B-Preview created with abliteration (see remove-refusals-with-transformers to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.

References

HuggingFace