9 Best Uncensored Local LLM (7 to 20B)

9 Min Read
New Project 14 - 9 Best Uncensored Local LLM (7 to 20B)

LLMs are awesome, but have you heard of uncensored ones? They blow the regular ones out of the water! Seriously, they can handle anything you throw at them, complex questions, wild roleplays, you name it. Plus, they’re way better at following instructions and having deep debates than the censored ones, which can be real sticklers sometimes.

In this article, I’m spilling the tea on the best uncensored LLMs out there, the ones with like 7 to 20 billion parameters (think super-powered brains). We’re gonna dive into what makes each one special, what they can do, and how you can actually use them in the real world. Basically, we’re about to unlock the true potential of chatty AI, no filters attached.

Related: 3 Open Source LLM With Longest Context Length

Contact me if you think some other model should be on the list.

Best Uncensored LLM Model

Check it out! We’re diving into some awesome open-source, uncensored language models. These aren’t your average chatbots – they’re powerful tools that put the control in your hands. Let’s see what makes each one unique!


fig 0 - 9 Best Uncensored Local LLM (7 to 20B)

So FuseChat, It’s like the ultimate chat buddy, uncensored and all. It’s basically three super-powered chatbots mashed into one, each with their own thing going on. That means you get crazy good conversations, no matter what you wanna shoot the breeze about.

Think you stumped it? Nah, FuseChat’s got your back. It’s like that super informed friend who always knows something cool. Plus, there’s this secret code (think “dolphin mode”) that unlocks its full brainpower for pretty much any question you can fire at it. Basically, FuseChat is your pocket AI sidekick for all things conversation!


Chimera-Apex-7B isn’t your grandma’s LLM. This one’s built to be uncensored and awesome for all sorts of stuff, from regular conversations to those, well, not-so-regular story ideas. Stuck on code? Need a brainstorming buddy who isn’t afraid to get a little wild? Chimera-Apex-7B’s your guy (or gal, it doesn’t judge).

Think of it as your super chill friend who can chat about literally anything, and we mean anything. It’s still under development, so expect some surprises, which is half the fun, right?


Let me tell you why the dolphin-2.8-experiment26-7b model is one of the best uncensored LLM models out there. This model is truly uncensored, meaning it can answer any question you throw at it, as long as you prompt it correctly. It’s like having a conversation with a knowledgeable friend who always has an answer for everything.

What sets this model apart is its ability to understand and respond to system prompts. When you use the system prompt of the dolphin model, you unlock its full potential. It’s like giving it a secret key that allows it to tap into its vast knowledge base and generate responses that are both accurate and insightful.

The dolphin-2.8-experiment26-7b model is a fine-tune of an experimental model that has proven to be the best 7 billion parameter model out there. It’s like the evolved version of an already impressive AI, with all the kinks ironed out and the performance optimized.

Using this model feels like having a personal AI assistant that’s always ready to help. It’s not just a tool, but a companion that makes your life easier. Whether you’re looking for answers to complex questions or just need some help with everyday tasks, the dolphin-2.8-experiment26-7b model is always up to the task.


The Nous-Hermes-2-Mistral-7B-DPO model, available on Hugging Face, represents a significant improvement, showcasing enhanced performance across various benchmarks compared to its predecessors. It is particularly noteworthy for its application in uncensored environments, offering a new level of engagement and interaction possibilities. This model’s design focuses on providing high-quality, synthetic data-driven responses, making it an excellent candidate for those seeking advanced, unrestricted LLM capabilities.


UNA-TheBeagle-7b-v1 is a top-notch, uncensored language model with 7 billion parameters. It’s trained on The Bagel dataset using Direct Preference Optimization (DPO) and UNA. It ranked #1 7b on the HF Leaderboard with an ARC score of 73. The model is based on Intel’s neural-chat model and performs well in many tasks. It’s available on the Hugging Face Model Hub.

Nous Hermes 2 – SOLAR 10.7B

I tested Nous Hermes 2 – SOLAR 10.7B, the newest Nous Research model based on the SOLAR 10.7B foundation. Trained on a massive dataset comprising mostly GPT-4 generated data and supplementary high-quality resources, this model excels in numerous benchmarks, nearly reaching the performance level of the Yi-34B model.

Notably, Nous Hermes 2 – SOLAR 10.7B’s system prompt capability, empowering users to define rules, roles, and even request uncensored responses from the model. This versatility enhances its applicability in roleplay, instruction following, and coding scenarios, placing it on par with many 30 billion parameter models.

Throughout my testing, I observed excellent results from this improved flagship model. Its uncensored response option adds an extra dimension to its functionality, broadening its appeal and potential uses.


Similar to the other large language models mentioned, DaturaCookie_7B is another uncensored LLM with 7 billion parameters. This model excels at roleplaying and lighthearted conversation, but be aware that it can also generate NSFW content.

Here’s a quick note for using DaturaCookie_7B’s vision capabilities: To leverage its multimodal features and incorporate vision into your tasks, you’ll need to load a specific mmproj file included within the model’s repository.

Dolphin 2.6 Mistral 7b – DPO Laser

I recently tried the Dolphin 2.6 Mistral 7b – DPO Laser LLM, an uncensored language model based on the LASER paper and enhanced by Fernando Fernandes and Ethan Hartford. With a larger context window of 16k tokens and advanced techniques like SVD noise reduction and RMT rank optimization, this model provides more robust outputs than its predecessors.

Personally, I found this uncensored model ideal for roleplay scenarios due to its wide response range. Its ability to reason impressively added depth to interactive simulations. Additionally, the removal of alignment and bias made it ethically sound yet highly adaptable to diverse user queries.

Overall, the Dolphin 2.6 Mistral 7b – DPO Laser LLM combines versatility, reliability, and creative freedom. As users, remember to employ an alignment layer before deployment and ensure ethical usage. Embrace the power of uncensored models while respecting ethical guidelines.


Dolphin-2.2.1-mistral-7b, developed by Eric Hartford and sponsored by a16z, is a remarkable open-source language model. Operating under the Apache-2.0 license, it is a versatile tool suitable for both commercial and non-commercial applications.

One of the standout features of Dolphin-2.2.1-mistral-7b is its commitment to fostering meaningful conversations and empathy. This model incorporates elements from curated Samantha DNA, enabling it to provide personalized advice and exhibit a genuine concern for the user’s emotions. Its capabilities have been further enhanced through extensive training in long multi-turn conversations.

An essential aspect of Dolphin-2.2.1-mistral-7b is its uncensored nature. The dataset has been meticulously filtered to eliminate any alignment and bias, ensuring that the model is more compliant and can provide a more neutral and open-ended approach to generating text. This commitment to neutrality and the absence of censorship makes Dolphin-2.2.1-mistral-7b an attractive choice for those looking to work with a versatile and adaptable open-source language model.

Share This Article
SK is a versatile writer deeply passionate about anime, evolution, storytelling, art, AI, game development, and VFX. His writings transcend genres, exploring these interests and more. Dive into his captivating world of words and explore the depths of his creative universe.