We are excited to introduce a powerful new feature for our platform—multiport support for custom image deployments!

The Qwen team has recently released the latest model series, Qwen 2.5, marking a significant advancement in their offerings. The model is designed to enhance performance, broaden application capabilities, and provide users with a more versatile AI experience.

LLama 3.3 is Meta's latest and most powerful 70-billion-parameter language model, offering state-of-the-art performance across a wide range of natural language processing tasks. It represents a significant step forward in terms of scalability, efficiency, and fine-tuning flexibility.

VLM finetuning allows you to adapt pre-trained multimodal models to better handle specific types of images and tasks relevant to your use case. Here are some key aspects:

microsoft/Phi-3-mini-4k-instruct and mistralai/Mistral-7B-Instruct-v0.2 serverless APIs will be deprecated on 18th January 2025. That means Phi3 mini and Mistral 7B LLM Serverless APIs will no longer process any requests from 18th January 2025 onwards.

We’re thrilled to announce a significant upgrade to our API! The Llama 3 8B model has now been upgraded to the latest Llama 3.1 8B version.

Unlock the Future of Language Models with Our Revolutionary Instruction Synthesizer API! 🚀

We're thrilled to unveil two major upgrades to MonsterTuner, designed to supercharge your LLM fine-tuning: Unsloth and Scaled Dot-Product Attention (SDPA). These innovations bring significant enhancements in performance, efficiency, and context length.

TinyLlama/TinyLlama-1.1B-Chat-v1.0 deprecated from July 3rd, 2024

We are pleased to announce the implementation of the latest model: google/gemma-2-9b-it.