jaahas/
qwen3-abliterated:14b

22.4K 4 months ago

This is the abliterated version of Qwen3, which is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.

tools 0.6b 1.7b 4b 8b 14b

4 months ago

f7b6f7c89fab · 9.0GB ·

model
qwen3
·
14.8B
·
Q4_K_M
template
{{- if .Messages }} {{- if or .System .Tools }}<|im_start|>system {{- if .System }} {{ .System }} {{
license
Apache License Version 2.0, January 2004 http://www.apache.org/licenses/ TERMS AND CONDITIONS FOR US
params
{ "repeat_penalty": 1, "stop": [ "<|im_start|>", "<|im_end|>" ], "te

Readme

Qwen 3 logo

The abliteration process for this model, especially the thinking mode, is still being adapted. Follow this thread for updates.

This model requires Ollama 0.6.6 or later. Download Ollama

Qwen 3 is the latest generation of large language models in the Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. The flagship model, Qwen3-235B-A22B, achieves competitive results in benchmark evaluations of coding, math, general capabilities, etc., when compared to other top-tier models such as Qwen3-30B-A3B and Qwen3-4B.

Key Features

  • Uniquely supports seamless switching between different models and configurations.
  • Significantly enhances performance and efficiency through advanced algorithms.
  • Superior human preference alignment for better user experience.
  • Supports 100+ languages for global accessibility.

Evaluation 1 Evaluation 2

Models

  • 0.6B parameter model

    ollama run jaahas/qwen3-abliterated:0.6b

  • 1.7B parameter model

    ollama run jaahas/qwen3-abliterated:1.7b

  • 4B parameter model

    ollama run jaahas/qwen3-abliterated:4b

  • 8B parameter model

    ollama run jaahas/qwen3-abliterated:8b

  • 14B parameter model

    ollama run jaahas/qwen3-abliterated:14b

  • 32B parameter model

    coming soon

  • 30B mixture-of-experts model with 3B active parameters

    ollama run jaahas/qwen3-abliterated:30b-a3b

  • 235B mixture-of-experts model with 22B active parameters

    coming soon

Reference