Changelog #80 - Qwen 2.5 VL 7B Instruct and Qwen 2.5 VL 72B Instruct One-Click Models, Improved Scale to Zero Cold Start, and more

Announcements
, , ,
1

Hello, and welcome to this week’s changelog update! Let’s dive into what’s new:

  1. Qwen 2.5 VL 7B Instruct and Qwen 2.5 VL 72B Instruct One-Click Models

    You can now deploy the latest Qwen 2.5 VL 7B Instruct and Qwen 2.5 VL 72B Instruct models with one click from the catalog.

    Qwen 2.5 VL 7B Instruct and Qwen 2.5 VL 72B Instruct One-Click Models
    Qwen 2.5 VL 7B Instruct and Qwen 2.5 VL 72B Instruct One-Click Models1920×1029 127 KB

  2. Improved Scale to Zero Cold Start

    When a scaled-to-zero service wakes up, we now prioritize reallocating it to the last hypervisor it ran on to reduce cold starts. Previously, the service could have been assigned to a different hypervisor, requiring a full container image re-pull.

  3. Best Open Source LLMs in 2025

    Looking for the best open-source LLM for your needs? Don’t miss our latest blog post that explores the top models available in early 2025, including DeepSeek-R1, Mistral Small 3, and Qwen 2.5 Coder 7B.