Text Generation Webui – v3.10 – Multimodal support!

📦 Text Generation Webui – v3.10 – Multimodal support!

🎉 Text Generation WebUI – v3.10 is here! 🎉

Big news for all you AI enthusiasts! The Text Generation WebUI has been updated to version 3.10 and it’s packed with some exciting new features, including Multimodal Support! That means you can now feed images into the LLM alongside text! 🤩

Here’s a breakdown of what’s new:

  • Multimodal Inputs: The UI & API now support images as input.
  • ExLlamaV3 Improvements: Speculative decoding is added, and it’s now the default loader for EXL3 models.
  • Chat Templates: Load custom chat templates from `chat_template.json` files!
  • API Boost: The API’s maximum token length has been increased to 512.

Plus a bunch of UI tweaks and bug fixes! Check out the release notes for all the details: https://github.com/oobabooga/text-generation-webui/releases/tag/v3.10

Grab the latest portable builds for Windows, Linux & Mac: https://github.com/oobabooga/text-generation-webui/releases/tag/v3.10

🔗 https://github.com/oobabooga/text-generation-webui/releases/tag/v3.10

AI
August 12, 2025 (0)
« »