DeepSeek's first generation reasoning models with comparable performance to OpenAI-o1.
435.1K Pulls 46 Tags Updated 7 weeks ago
QwQ is an experimental research model focused on advancing AI reasoning capabilities.
55.7K Pulls 17 Tags Updated 2 weeks ago
Ablitered v3 llama-3.1 8b with uncensored prompt
20.8K Pulls 38 Tags Updated 7 months ago
Qwen2.5-1M is the long-context version of the Qwen2.5 series models, supporting a context length of up to 1M tokens.
18.1K Pulls 11 Tags Updated 7 weeks ago
This is a phi3 medium model that has had the prompt rejection neurons snipped which is better than fine tuning as it doesn't change how the model acts
5,925 Pulls 1 Tag Updated 10 months ago
Abliterated version of Google's Gemma 2 9B
5,545 Pulls 1 Tag Updated 8 months ago
mlabonne/NeuralDaredevil-8B-abliterated
4,978 Pulls 4 Tags Updated 7 months ago
3,545 Pulls 1 Tag Updated 9 months ago
New state of the art 70B model. Llama 3.3 70B offers similar performance compared to Llama 3.1 405B model.
3,421 Pulls 10 Tags Updated 3 months ago
NeuralDaredevil-8B-abliterated with more quantization
3,318 Pulls 5 Tags Updated 9 months ago
Phi-4 is a 14B parameter, state-of-the-art open model from Microsoft.
3,005 Pulls 5 Tags Updated 2 months ago
llama3.2:3b abliterated (aka. uncensored)
2,802 Pulls 4 Tags Updated 4 months ago
From huihui-ai/Llama-3.2-11B-Vision-Instruct-abliterated
2,470 Pulls 2 Tags Updated 7 weeks ago
Many quantized GGUF versions of deepseek R1 abliterated (uncensored) with tools support
2,209 Pulls 8 Tags Updated 7 weeks ago
Mistral Small 3 sets a new benchmark in the “small” Large Language Models category below 70B.
1,980 Pulls 10 Tags Updated 7 weeks ago
Contains fp16 version of the excellent NeuralDaredevil-8B-abliterated
1,827 Pulls 1 Tag Updated 8 months ago
Qwen2.5 7B, but Josiefied and uncensored model
1,659 Pulls 1 Tag Updated 5 months ago
This is a model that has had the neurons for refusal snipped, allowing for a better uncensored model without the need for fine-tuning that might otherwise effect its behavior
1,576 Pulls 1 Tag Updated 10 months ago
The mradermacher model of DeepSeek-r1 llama distilled with abliteration applied
1,330 Pulls 1 Tag Updated 2 months ago
The Skywork o1 Open model series, developed by the Skywork team at Kunlun Inc. This groundbreaking release introduces a series of models that incorporate o1-like slow thinking and reasoning capabilities.
1,267 Pulls 5 Tags Updated 3 months ago