EN

FREE DeepSeek Is BACK! New FREE Proxy Is INSANE! (BYE Character.AI & NO FILTERS)

#AI Tools
2025-11-28 14:138 min read

Content Introduction
Ask Questions

In this tutorial video, the host addresses role-playing enthusiasts who are dissatisfied with Character.AI's limitations and slow responses. They introduce free AI models including Deepseek and GLM 4.6, which promise better performance and flexibility. The video provides a step-by-step guide on how to set up these models using janitor.ai, including creating API keys and configuring settings for optimal role-playing experiences. The host emphasizes the importance of selecting the right models and offers tips on generating responses. Additional features such as message limits and context sizes are discussed, along with troubleshooting advice for common issues. Viewers are encouraged to subscribe for more tutorials and AI setups, especially for character AI alternatives, as the host concludes with a friendly reminder to enjoy role-playing.

Key Information

  • The video introduces free Deepseek models and the new version GLM4.6, emphasizing their utility for role players tired of Character.AI's limitations.
  • This video guides users on accessing various free models, including GPT4.1 and others, detailing usage of a proxy providing 200 free messages daily.
  • Users can connect via janitor.ai, with instructions provided for model selection, configuration, and advanced settings.
  • It highlights the importance of setting optimal parameters like temperature and context size to enhance roleplay experience.
  • The creator encourages viewer participation by liking and subscribing for future tutorials and LLM setups.

Timeline Analysis

Content Keywords

Deepseek Models

The video introduces free Deepseek models, focusing on their advantages over Character.AI's filters and slow responses, presenting them as viable alternatives for roleplayers.

GLM4.6

GLM4.6 is highlighted as a free model with a strong performance, rivaling Deepseek. The video details its features, including a low hallucination rate, making it suitable for roleplay.

Proxy Features

A proxy giving users 200 free messages daily is presented, offering a better option compared to existing routers, with instructions on how to access and utilize it effectively.

Chub.ai

Chub.ai is mentioned as a platform that may require additional configurations to integrate effectively, with a note on covering its complexities in future videos.

Prompt Configuration

The script guides viewers on configuring prompts for these AI models, including creating API keys and adjusting settings for optimal roleplay experience.

Advanced Model Settings

Details regarding advanced settings for each model are shared, including recommendations for temperature and context size, enhancing the user’s experience.

Troubleshooting

The video advises on refreshing the site for errors and gives tips on handling proxy issues, ensuring users have a smooth experience while using the models.

Roleplaying Enhancements

The video encourages roleplayers to explore these alternatives and enhancements for a better gaming experience, emphasizing the customization available in new AI models.

YouTube Channel Subscription

Viewers are prompted to like and subscribe for more tutorials, free LLM setups, and character AI alternatives, fostering engagement with the content.

Related questions&answers

What is the main purpose of the video?

The video introduces free Deepseek models and explains how to use them for roleplaying.

What are some of the features of the new GLM4.6 model?

GLM4.6 has 203K context, approximately 357B parameters, and a very low hallucination rate, which is great for roleplay.

How many messages can I send daily with the provided proxy?

The proxy allows for 200 free messages daily.

What should I do if I reach the 20 messages per minute limit?

If you hit the 20 messages per minute limit, you'll be rate limited for about 20 to 30 seconds.

How do I set up a new configuration in janitor.ai?

To set up a new configuration, enter any chat, press the three lines icon, select 'Add configuration', and name it. Then, input the model name and proxy URL.

What does '0/0' signify when choosing a model?

'0/0' indicates that the model is a fresh selection without any prior limits.

What settings should I adjust for optimal performance?

For temperature, a range between 0.8 and 1.2 is recommended, and for best roleplaying memory, set the context size to 64,000.

What should I do if I encounter network errors?

If you see network errors, refresh the page a few times, as it may be an issue with janitor.

How can I edit the prompt I use for the models?

You can edit the prompt by going to chub.ai, typing 'deep clean', and modifying the entire system prompt as necessary.

Is there a way to get updates or more tutorials?

Yes, you can stay tuned for more tutorials and updates by subscribing to the channel.

More video recommendations

Share to: