Model reference · open weights
Supra2 is an open-weight language model from SupraLabs. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.
Available as managed deployment — configured and operated for you on dedicated EU hardware, quoted per deployment.
What it is
| Released by | SupraLabs |
|---|---|
| Type | Language models |
| Task | Text gen |
| Parameters (lead) | 101M |
| Context | 2k tokens |
| Runs with | transformers |
| Based on | SupraLabs/Supra2-100M |
| Released | 2026-08-03 |
| Popularity | 26k downloads / month |
| Licence | Open weights |
About
Instruct model of Supra2-100M • 100M Parameters • 2K Context
Supra2-100M Instruct is a 100M-parameter decoder-only language model pretrained from scratch by SupraLabs on 30B tokens of English web text. It uses the Qwen3 architecture with a custom 32,768-token tokenizer. This is the instruct version of Supra2-100M Base and it was full-finetuned on a single RTX 5060 Ti 16GB.
Prompt: "What is AI?" Answer:
AI refers to the application of computer science and artificial intelligence to solve problems and make decisions. It's a field that combines computer science, machine learning, and natural language processing to create new forms of intelligence, such as natural language understanding, speech recognition, and virtual assistants.
AI is used in various fields, including healthcare, finance, education, and social sciences, to improve efficiency, reduce costs, and enhance human performance. It can also be applied to areas like robotics, artificial intelligence, and cybersecurity. In the context of AI, it's often used to analyze vast amounts of data, identify patterns, and make predictions. For example, in natural language analysis, AI can help humans understand and interpret language, such that they can make more accurate and informed decisions. The field of AI is constantly evolving, with new technologies emerging and being developed to address specific challenges. Some examples include:
Artificial intelligence (AI) is being used in robotics, where it can learn from human movement and adapt to changing environments. AI-powered virtual assistants, such virtual assistants that can help users interact with machines, such Aspire and Alexa, are being developed. Artificially intelligent vehicles, such vehicles that can learn and adapt based on user input, are also being developed, such cars and trucks. Smart home systems, such smart thermostats and smart lighting, are becoming increasingly sophisticated and integrated into homes. Autonomous vehicles, like autonomous cars and drones, are increasingly used in urban areas and in remote areas. These are just a few examples, but the field of artificial intelligence is constantly expanding and evolving, and AI is being applied to a wide range of applications.
Followup prompt: "Can you give me more information about how it will influence work life?" --> Answer:
The impact of AI on work life will depend on several factors, including the type of work, the level of automation, and the level and type of tasks being performed. Here are some key aspects to consider: Tasks being performed:
- Work-life balance: Will AI-driven tasks lead to increased productivity, creativity, and overall well-being?
- Machine learning algorithms*: How will AI-powered algorithms adapt to new situations and tasks?
Level of automation: Can AI-generated tasks be automated, and will they be more effective than manual labor? *** Robustness: Are there any potential risks or downsides to AI-based tasks, such tasks that require human judgment or expertise?
- Sensor and analytics: What are the potential risks and benefits of AI-assisted tasks, and how will they impact work-life balances? *** Automation: Does AI-led automation have the potential to automate repetitive tasks, improve workflows, and reduce errors?
- Competition for jobs: Is there a significant gap between the demand for AI-related jobs and the supply of skilled workers? By incorporating AI into work, we can create more efficient, effective, and sustainable work environments. This can lead to improved employee well-Being, reduced stress, and increased productivity.
Followup prompt: "Thank you." --> Answer:
I'm glad I could help. I'd be happy to help you with any questions or concerns you may have. Please go ahead and share your work, and I'll do my best to assist you in understanding the implications of AI in your work.
Prompt: "What is google?" Answer:
Google is a web-based platform that allows users to search and find information on the web. It's a social media platform that uses algorithms to make recommendations based on various factors such as location, time of day, and interests.
Google has a number of features that make it easy to find information, including:
- Searching for keywords and phrases related to various topics, like books, movies, or music
- Analyzing website traffic and traffic patterns
- Creating a custom search interface
- Suggesting alternative ways to find the information
- Providing recommendations for books, articles, and other content
- Allowing users to customize the search results
One of the main advantages of Google is that it's easy to use, as users can search for the information they need, and then filter the results based on their interests. This makes it accessible to a wider audience.
Followup prompt: "How does the traffic analyzation work?" --> Answer:
The traffic analysis process involves several steps to understand the flow of traffic to and from the website. Here's a breakdown of the steps involved:
Searching: Users will typically use a browser to search for relevant websites. The site will be crawled by a server and the results will be displayed in a search box.
Scheduling: The website will be scheduled to run on the client machine, which is usually a mobile device like a tablet or smartphone. The system will generate a list of potential candidates and will then prioritize them based on criteria such as relevance, interest, and user engagement.
Optimization: Once the data is analyzed, it will be aggregated into a map that showcases the most popular pages, pages with the highest traffic, and the most traffic-related pages. This will give the website an idea of the traffic patterns and help the user make informed decisions about the search.
Marketing: Th
From the published model card. Full card on the HuggingFace links in the sidebar.
Using it via the API
Once AxForge deploys supra2 for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (supra2 below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/chat/completions \
-H "Authorization: Bearer $AXFORGE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"supra2","messages":[{"role":"user","content":"Hello"}]}'
Create an account — your API key is available in the console. 3M free tokens every 30 days with every new account.