Download Latest Version textgen-portable-3.21-macos-arm64.zip (198.5 MB)
Email in envelope

Get an email when there's a new version of Text Generation Web UI

Home / v3.19
Name Modified Size InfoDownloads / Week
Parent folder
textgen-portable-3.19-windows-cuda12.4.zip 2025-11-29 910.5 MB
0
textgen-portable-3.19-windows-vulkan.zip 2025-11-29 213.8 MB
0
textgen-portable-3.19-windows-cpu.zip 2025-11-29 199.5 MB
0
textgen-portable-3.19-linux-cuda12.4.zip 2025-11-29 1.6 GB
0
textgen-portable-3.19-linux-rocm.zip 2025-11-29 618.4 MB
0
textgen-portable-3.19-macos-x86_64.zip 2025-11-29 166.4 MB
0
textgen-portable-3.19-linux-vulkan.zip 2025-11-29 293.2 MB
0
textgen-portable-3.19-linux-cpu.zip 2025-11-29 249.6 MB
0
textgen-portable-3.19-macos-arm64.zip 2025-11-29 192.9 MB
0
README.md 2025-11-29 1.2 kB
0
v3.19 source code.tar.gz 2025-11-29 24.9 MB
0
v3.19 source code.zip 2025-11-29 25.0 MB
0
Totals: 12 Items   4.5 GB 0

Qwen3-Next llama.cpp support!

Changes

  • Add slider for --ubatch-size for llama.cpp loader, change defaults for better MoE performance (#7316). Thanks, @GodEmperor785.
  • This significantly improves prompt processing speeds for MoE models in both full-GPU and GPU+CPU configurations.

Bug fixes

  • fix(deps): upgrade coqui-tts to >=0.27.0 for transformers 4.55 compatibility (#7329). Thanks, @aidevtime.

Backend updates

Portable builds

Below you can find self-contained packages that work with GGUF models (llama.cpp) and require no installation! Just download the right version for your system, unzip, and run.

Which version to download:

  • Windows/Linux:
  • NVIDIA GPU: Use cuda12.4.
  • AMD/Intel GPU: Use vulkan builds.

  • CPU only: Use cpu builds.

  • Mac:

  • Apple Silicon: Use macos-arm64.

Updating a portable install:

  1. Download and unzip the latest version.
  2. Replace the user_data folder with the one in your existing install. All your settings and models will be moved.
Source: README.md, updated 2025-11-29