Join the event trusted by corporate leaders for almost two decades. VB Transforma connects people building AI Real Enterprise. Learn more
Monster tokenThe modern AI Chatbot platform has introduced ALFA preview, aimed at changing the interaction of users with enormous models (LLM).
Developed by Matt Shumer, co -founder and CEO Other and his hit Ai Writing Assistant Hipwrite AIThe key point of sale of the Monster token is its ability to conduct the user to the best available LLM for a given task, providing improved results by using the strength of many models.
Seven main LLM are available via the monster token. When the user enters something in the jumping field, the Monster token uses preliminary compartments developed through iteration by Shumer himself to automatically analyze the user’s input details, decide which combination of many available models and related tools is best suited for response, and then ensure a total reaction using the strengths of the mentioned models. Available LLM includes:
- Anthropic Claude Sonet 3.5
- Anthropic claude 3.5 opus
- Openai GPT-4.1
- Openai GPT-4O
- AI PPLX embarrassment (for testing)
- OpenAI O3 (for reasoning)
- Google Gemini 2.5 Pro
Unlike other Chatbot platforms, the Monster token automatically identifies which LLM is the best for specific tasks-which tools connected to LLM would be helpful, such as website search circles or coding-I organizes work flow in many model.
“We just build connectors for everything, and then a system that decides what to use when,” said Shumer.
For example, it can employ, among others, Claude for creativity, O3 for reasoning and PPLX. This approach eliminates the need to manually choose the right model for each poem, simplifying the process for anyone who wants high quality, adapted results.
Backlighting of the function
Preview Alpha, which can now be saved to tokenmonster.AI, allows users to send the scope of file types, including Excel, PowerPoint and Docs.
It also contains functions such as website extraction, lasting conversation sessions and “fast mode”, which automatically leads to the best model without entering the user.
OpenRoute is in the heart of token monster, an external service that acts like a gate to many LLM, and in which Shumer has invested a petite sum at its party.
This architecture allows the Monster token to employ a number of models from various suppliers without building separate integration for each of them.
Prices and availability
At the moment, the Monster token does not charge a flat monthly fee.
Instead, users only pay for tokens that consume via OpenRouter, which makes it adaptable for different levels of employ.
According to Shumer, this model has been inspired by Cline, a tool that allows high -earning users access to unlimited AI Power, enabling them to achieve better results, simply using more computing resources.
Multi -stage work flows produce richer LLM answers
AI token monster work boards go beyond straightforward brisk routing.
In one example, Chatbot can start with the research phase using the Internet Search API interfaces, transfer this data to O3 to identify gaps in information, and then create an outline from Gemini 2.5 Pro, sketch of the text with Claude Opus and improve it with the Claude 3.5 sonnet.
This multi -stage orchestration has been designed to provide richer, fuller answers than a single LLM can be able to generate alone.
The platform also includes the ability to save sessions, and the data is securely stored using the Open Source database internet database service Supabase. This ensures that users can return to ongoing projects without losing their jobs, while giving them control over saved data and ephemeral.
Non -traditional CEO
In a noteworthy experiment, the leadership of the token monster was handed over to the Claude Anthropica model.
Shumer announced that he is obliged to make any decision made by “Claude CEO”, calling it a test to check if AI can effectively manage the company.
“Either we revolutionized management forever or we made a huge mistake,” he wrote on X.
Emerging from the reflection of 70-b controversy
The premiere of the Monster tokena appears less than a year after Shumer faced the controversy in connection with its launch and the final withdrawal of the 70b reflection, the refined version of Llam 3.1 Meta Llama 3.1, which was initially advertised as the most perfect Open Source model in the world, but quickly became the subject of criticism and accusations of fraud after it could not be played outside, but which quickly became a prevailing test.
Shumer apologized and said that the problems were born of mistakes made because of speed. The episode emphasized the challenges and risk of rapid development of artificial intelligence and the importance of transparency in model versions.
MCP integrations are coming next
Shumer said that his team on the Monster token also examines modern possibilities, such as integration with the Model Context Protocol (MCP) servers, which allow websites and companies to have the employ of their knowledge, tools and products to achieve higher -order tasks than just text or image generation.
This would allow the Monster token to combine with the user’s internal data and services, opening the possibilities to handle tasks such as managing customer service tickets or interface with other business systems.
Shumer emphasized that the Monster token is still very at an early stage. Although it already supports a package of powerful functions, the platform remains ALFA and is expected to see brisk iterations and updates, because more users provide opinions. “We will continue to harden and add things,” he said.
Promising experiment
For users who want to take advantage of the connected power of many LLM without any problems with switching the model, the Monster token can be an attractive choice. It is designed for work for people who do not want to spend hours upgrading hints or testing various models independently, instead allowing the system to automate routing and multi -stage work flows to handle complexity.
With the boost in the possibility of the Monster tokena, it will be intriguing to see how users and companies accept to-i like his experiment with management managed by AI. For now, it is a promising addition to the rapidly developing landscape of AI chatbots and digital assistants.
