FAQ

​

1. Is Prem free to use?

• Yes, Prem is free, and for premium features, you can choose between different pricing tiers.

Check out our pricing page to learn more.

​

2. What models are available to use with Prem?

The following models are available to use with Prem:

​

Anthropic

• claude-3-5-sonnet
• claude-3.5-sonnet-v2
• claude-3.5-haiku
• claude-3-haiku
• claude-3-opus
• claude-3-sonnet

​

codellama

• prem-codellama-7b

​

Deepseek

• deepseek-r1
• deepseek-r1-distill-llama-70b

​

Gemma2

• prem-gemma2-2b
• prem-gemma2-9b

​

GPT

• gpt-4o-mini
• gpt-4o-mini-eu
• gpt-4o
• gpt-4o-eu
• gpt-4-eu
• gpt-4-turbo

​

Llama

• llama-3.3-70b
• llama-3.2-1b
• llama-3.2-3b
• llama-3.1-8b
• llama-3.1-70b
• llama-3-8b-guard
• llama-3-8b
• llama-3-70b
• llama-3-8b-guard
• prem-llama-3.2-1b
• prem-llama-3.2-3b
• prem-llama-3.1-8b

​

Mistral

• mistral-8x7b-instruct

​

Qwen

• prem-qwen-2.5-3b
• prem-qwen-2.5-7b
• prem-qwen-2.5-0.5b
• prem-qwen-2.5-1.5b

​

3. Is there an API that I can use if I am not using Python or Javascript?

• Yes, our API is language-agnostic, which means you can use it with any programming language that supports making HTTP requests. While our documentation provides examples in Python and JavaScript SDKs, you can easily adapt these examples to work with your preferred programming language, such as GO, Java, Rust, or PHP. As long as your chosen language has a built-in way to send HTTP requests and parse JSON responses, you can integrate our API into your projects seamlessly.

​

4. How does the Autonomous fine-tuning Agent work? Do I need to be an experienced ML engineer?

• You do not need to be an experienced ML engineer to use the Autonomous fine-tuning Agent. All you need to do is provide a dataset and chooose a base model. The Agent will then fine-tune the model on your dataset. Check out our docs to learn more.

​

5. How do I migrate projects between organizations?

• At present, we do not offer assistance for migrating projects between organizations. While you can replicate this process manually by utilizing the SDK to read and write runs. the most efficient approach would be to create a new project within your organization and proceed from there.