Changes

• log error when llama-server request exceeds context size (#7263). Thanks, @mamei16.
• Make --trust-remote-code immutable from the UI/API for better security.

Bug fixes

• Fix metadata leaking into branched chats.
• Fix "continue" missing an initial space in chat-instruct/chat modes.
• Fix resuming incomplete downloads after HF moved to Xet.
• Revert exllamav3_hf changes in v3.14 that made it output gibberish.

Backend updates

• Update llama.cpp to https://github.com/ggml-org/llama.cpp/tree/f9fb33f2630b4b4ba9081ce9c0c921f8cd8ba4eb.
• Update exllamav3 0.0.10.

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 for newer GPUs or cuda11.7 for older GPUs and systems with older drivers.
  • AMD/Intel GPU: Use vulkan builds.
  • CPU only: Use cpu builds.

• Mac:

  • Apple Silicon: Use macos-arm64.
  • Intel CPU: Use macos-x86_64.

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.

Changes

• Better handle multi-GPU setups when using Transformers with bitsandbytes (load-in-8bit and load-in-4bit).
• Implement the /v1/internal/logits endpoint for the exllamav3 and exllamav3_hf loaders.
• Make profile picture uploading safer.
• Add fla to the requirements for Exllamav3 to support qwen3-next models.

Bug fixes

• Fix an issue with loading certain chat histories in Instruct mode. Thanks, @Remowylliams.
• Fix portable builds for macOS x86 missing llama.cpp binaries (#7238). Thanks, @IonoclastBrigham.

Backend updates

• Update llama.cpp to https://github.com/ggml-org/llama.cpp/tree/d00cbea63c671cd85a57adaa50abf60b3b87d86f.
• Update transformers to 4.57.
• Update exllamav3 0.0.7.
• Update bitsandbytes to 0.48.

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 for newer GPUs or cuda11.7 for older GPUs and systems with older drivers.
  • AMD/Intel GPU: Use vulkan builds.
  • CPU only: Use cpu builds.

• Mac:

  • Apple Silicon: Use macos-arm64.
  • Intel CPU: Use macos-x86_64.

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.

Bug fixes

• Don't use $ $ for LaTeX, only $$ $$, to avoid broken rendering of text like apples cost $1, oranges cost $2
• Fix exllamav3 ignoring the stop button
• Fix a transformers issue when using --bf16 and Flash Attention 2 (#7217). Thanks, @stevenxdavis.
• Fix x86_64 macos portable builds containing arm64 files

Backend updates

• Update llama.cpp to https://github.com/ggml-org/llama.cpp/tree/7f766929ca8e8e01dcceb1c526ee584f7e5e1408
• Update transformers to 4.56

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 for newer GPUs or cuda11.7 for older GPUs and systems with older drivers.
  • AMD/Intel GPU: Use vulkan builds.
  • CPU only: Use cpu builds.

• Mac:

  • Apple Silicon: Use macos-arm64.
  • Intel CPU: Use macos-x86_64.

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.

Changes

• Characters can now think in chat-instruct mode! This was possible thanks to many simplifications and improvements to jinja2 template handling:

• Add support for the Seed-OSS-36B-Instruct template.
• Better handle the growth of the chat input textarea:

Before	After

• Make the --model flag work with absolute paths for gguf models, like --model /tmp/gemma-3-270m-it-IQ4_NL.gguf
• Make venv portable installs work with Python 3.13
• Optimize LaTeX rendering during streaming for long replies
• Give streaming instruct messages more vertical space
• Preload the instruct and chat fonts for smoother startup
• Improve right sidebar borders in light mode
• Remove the --flash-attn flag (it's always on now in llama.cpp)
• Suppress "Attempted to select a non-interactive or hidden tab" console warnings, reducing the UI CPU usage during streaming
• Statically link MSVC runtime to remove the Visual C++ Redistributable dependency on Windows for the llama.cpp binaries
• Make the llama.cpp terminal output with --verbose less verbose

Bug fixes

• llama.cpp: Fix stderr deadlock while loading some models
• llama.cpp: Fix obtaining the maximum sequence length for GPT-OSS
• Fix the UI failing to launch if the Notebook prompt is too long
• Fix LaTeX rendering for equations with asterisks
• Fix italic and quote colors in headings

Backend updates

• Update llama.cpp to https://github.com/ggml-org/llama.cpp/tree/9961d244f2df6baf40af2f1ddc0927f8d91578c8

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 for newer GPUs or cuda11.7 for older GPUs and systems with older drivers.
  • AMD/Intel GPU: Use vulkan builds.
  • CPU only: Use cpu builds.

• Mac:

  • Apple Silicon: Use macos-arm64.
  • Intel CPU: Use macos-x86_64.

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.

Changes

• Add the Tensor Parallelism option to the ExLlamav3/ExLlamav3_HF loaders through the --enable-tp and --tp-backend options.
• Set multimodal status during Model Loading instead of checking every generation (#7199). Thanks, @altoiddealer.
• Improve the multimodal API examples slightly.

Bug fixes

• Make web search functional again
• mtmd: Fix a bug when "include past attachments" is unchecked
• Fix code blocks having an extra empty line in the UI

Backend updates

• Update llama.cpp to ggml-org/llama.cpp@6d7f111
• Update ExLlamaV3 to 0.0.6
• Update flash-attention to 2.8.3

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 for newer GPUs or cuda11.7 for older GPUs and systems with older drivers.
  • AMD/Intel GPU: Use vulkan builds.
  • CPU only: Use cpu builds.

• Mac:

  • Apple Silicon: Use macos-arm64.
  • Intel CPU: Use macos-x86_64.

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.

See the Multimodal Tutorial

Changes

• Add multimodal support to the UI and API
  • With the llama.cpp loader (#7027). This was possible thanks to PR ggml-org/llama.cpp#15108 to llama.cpp. Thanks @65a.
  • With ExLlamaV3 through a new ExLlamaV3 loader (#7174). Thanks @Katehuuh.
• Add speculative decoding to the new ExLlamaV3 loader.
• Use ExLlamav3 instead of ExLlamav3_HF by default for EXL3 models, since it supports multimodal and speculative decoding.
• Support loading chat templates from chat_template.json files (EXL3/EXL2/Transformers models)
• Default max_tokens to 512 in the API instead of 16
• Better organize the right sidebar in the UI
• llama.cpp: Pass --swa-full to llama-server when streaming-llm is checked to make it work for models with SWA.

Bug fixes

• Fix getting the ctx-size for newer EXL3/EXL2/Transformers models
• Fix the exllamav2 loader ignoring add_bos_token
• Fix the color of italic text in chat messages
• Fix edit window and buttons in Messenger theme (#7100). Thanks @mykeehu.

Backend updates

• Bump llama.cpp to ggml-org/llama.cpp@f4586ee

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 for newer GPUs or cuda11.7 for older GPUs and systems with older drivers.
  • AMD/Intel GPU: Use vulkan builds.
  • CPU only: Use cpu builds.

• Mac:

  • Apple Silicon: Use macos-arm64.
  • Intel CPU: Use macos-x86_64.

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.

Changes

• Several improvements to the GPT-OSS template handling. Special actions like "Continue" and "Impersonate" now work correctly.
• Update llama.cpp to https://github.com/ggml-org/llama.cpp/tree/5fd160bbd9d70b94b5b11b0001fd7f477005e4a0

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 for newer GPUs or cuda11.7 for older GPUs and systems with older drivers.
  • AMD/Intel GPU: Use vulkan builds.
  • CPU only: Use cpu builds.

• Mac:

  • Apple Silicon: Use macos-arm64.
  • Intel CPU: Use macos-x86_64.

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.

Experimental GPT-OSS support!

I have obtained some success with the GGUF models under

https://huggingface.co/ggml-org/gpt-oss-20b-GGUF/tree/main
https://huggingface.co/ggml-org/gpt-oss-120b-GGUF/tree/main

It may be necessary to re-download those models in the next days if bugs are found, so make sure to recheck those pages.

Changes

• Add a new Reasoning effort UI element in the chat tab, with low, medium, and high options for GPT-OSS
• Support standalone .jinja chat templates -- makes it possible to load GPT-OSS through Transformers
• Make web search functional with thinking models

Bug fixes

• Fix an edge case in chat history loading that caused a crash (closes #7155)
• Handle both int and str types in grammar char processing (fixes a rare crash when using grammar)

Backend updates

• Update llama.cpp to https://github.com/ggml-org/llama.cpp/tree/fd1234cb468935ea087d6929b2487926c3afff4b
• Update Transformers to 4.55 (adds GPT-OSS support)

Portable builds

Below you can find portable builds: 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 for newer GPUs or cuda11.7 for older GPUs and systems with older drivers.
  • AMD/Intel GPU: Use vulkan builds.
  • CPU only: Use cpu builds.

• Mac:

  • Apple Silicon: Use macos-arm64.
  • Intel CPU: Use macos-x86_64.

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.

Changes

• Replace use_flash_attention_2/use_eager_attention with a unified attn_implementation in the Transformers loader
• Ignore add_bos_token in instruct prompts, let the jinja2 template decide
• Add a "None" option for the speculative decoding model

Backend updates

• Update llama.cpp to https://github.com/ggml-org/llama.cpp/tree/90083283ec254fa8d33897746dea229aee401b37
• Update Transformers to 4.53
  • Also update bitsandbytes/Accelerate/PEFT to the latest versions
• Update ExLlamaV3 to 0.0.5
• Update ExLlamaV2 to 0.3.2

Portable builds

Below you can find portable builds: 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 for newer GPUs or cuda11.7 for older GPUs and systems with older drivers.
  • AMD/Intel GPU: Use vulkan builds.
  • CPU only: Use cpu builds.

• Mac:

  • Apple Silicon: Use macos-arm64.
  • Intel CPU: Use macos-x86_64.

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.

Changes

• Chat tab improvements:
  • Move the 'Enable thinking' checkbox from the Parameters tab to the right sidebar
  • Keep the last chat message visible as the input area grows
  • Optimize chat scrolling again (I think that will be the last time—it's really responsive now)
  • Replace 'Generate' with 'Send' in the main button
• Support installing user extensions in user_data/extensions/ for convenience
• Small UI optimizations and style improvements
• Block model and session backend events in --multi-user mode (#7098). Thanks @Alidr79
• One-click installer: Use miniforge instead of miniconda to avoid Anaconda licensing issues for organizations with 200+ people
• Standardize margins and paddings across all chat styles (new in 3.7.1)
• Update the keyboard shortcuts documentation (new in 3.7.1)
• docs: Add Mirostat Explanation (#7128). Thanks @Cats1337. (new in 3.7.1)

Bug fixes

• Fix the DuckDuckGo search
• Fix scrolling during streaming when thinking blocks are present
• Fix chat history getting lost if the UI is inactive for a long time
• Fix chat sidebars toggle buttons disappearing (#7106). Thanks @philipp-classen
• Fix autoscroll after initial fonts loading
• Handle either missing <think> start or </think> end tags (#7102). Thanks @zombiegreedo
• Fix custom stopping strings being reset when switching models
• Fix navigation icons temporarily hiding when switching message versions (new in 3.7.1)
• Revert "Keep the last chat message visible as the input area grows", as it was very glitchy (new in 3.7.1)

Backend updates

• Bump llama.cpp to ggml-org/llama.cpp@6491d6e

Portable builds

Below you can find portable builds: 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 for newer GPUs or cuda11.7 for older GPUs and systems with older drivers.
  • AMD/Intel GPU: Use vulkan builds.
  • CPU only: Use cpu builds.

• Mac:

  • Apple Silicon: Use macos-arm64.
  • Intel CPU: Use macos-x86_64.

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.