[
  {
    "version": "2.2.2",
    "notes":
"
* repeat penalty for both gptj and llama models
* scroll the context window when conversation reaches context limit
* persistent thread count setting
* new default template
* new settings for model path, repeat penalty
* bugfix for settings dialog onEditingFinished
* new tab based settings dialog format
* bugfix for datalake when conversation contains forbidden json chars
* new C library API and split the backend into own separate lib for bindings
* apple signed/notarized dmg installer
* update llama.cpp submodule to latest
* bugfix for too large of a prompt
* support for opt-in only anonymous usage and statistics
* bugfixes for the model downloader and improve performance
* various UI bugfixes and enhancements including the send message textarea automatically wrapping by word
* new startup dialog on first start of a new release displaying release notes and opt-in buttons
* new logo and icons
* fixed apple installer so there is now a symlink in the applications folder
",
    "contributors":
"
* Adam Treat (Nomic AI)
* Aaron Miller
* Matthieu Talbot
* Tim Jobbins
* chad (eachadea)
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.3.0",
    "notes":
"
* repeat penalty for both gptj and llama models
* scroll the context window when conversation reaches context limit
* persistent thread count setting
* new default template
* new settings for model path, repeat penalty
* bugfix for settings dialog onEditingFinished
* new tab based settings dialog format
* bugfix for datalake when conversation contains forbidden json chars
* new C library API and split the backend into own separate lib for bindings
* apple signed/notarized dmg installer
* update llama.cpp submodule to latest
* bugfix for too large of a prompt
* support for opt-in only anonymous usage and statistics
* bugfixes for the model downloader and improve performance
* various UI bugfixes and enhancements including the send message textarea automatically wrapping by word
* new startup dialog on first start of a new release displaying release notes and opt-in buttons
* new logo and icons
* fixed apple installer so there is now a symlink in the applications folder
* fixed bug with versions
* fixed optout marking
",
    "contributors":
"
* Adam Treat (Nomic AI)
* Aaron Miller
* Matthieu Talbot
* Tim Jobbins
* chad (eachadea)
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.4.0",
    "notes":
"
* reverse prompt for both llama and gptj models which should help stop them from repeating the prompt template
* resumable downloads for models
* chat list in the drawer drop down
* add/remove/rename chats
* persist chats to disk and restore them with full context (WARNING: the average size of each chat on disk is ~1.5GB)
* NOTE: to turn on the persistent chats feature you need to do so via the settings dialog as it is off by default
* automatically rename chats using the AI after the first prompt/response pair
* new usage statistics including more detailed hardware info to help debug problems on older hardware
* fix dialog sizes for those with smaller displays
* add support for persistent contexts and internal model state to the C api
* add a confirm button for deletion of chats
* bugfix for blocking the gui when changing models
* datalake now captures all conversations when network opt-in is turned on
* new much shorter prompt template by default
",
    "contributors":
"
* Adam Treat (Nomic AI)
* Aaron Miller
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.4.1",
    "notes":
"
* compress persistent chats and save order of magnitude disk space on some small chats
* persistent chat files are now stored in same folder as models
* use a thread for deserializing chats on startup so the gui shows window faster
* fail gracefully and early when we detect incompatible hardware
* repeat penalty restore default bugfix
* new mpt backend for mosaic ml's new base model and chat model
* add mpt chat and base model to downloads
* lower memory required for gptj models by using f16 for kv cache
* better error handling for when a model is deleted by user and persistent chat remains
* add a user default model setting so the users preferred model comes up on startup
",
    "contributors":
"
* Adam Treat (Nomic AI)
* Zach Nussbaum (Nomic AI)
* Aaron Miller
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.4.2",
    "notes":
"
* add webserver feature that offers mirror api to chatgpt on localhost:4891
* add chatgpt models installed using openai key to chat client gui
* fixup the memory handling when switching between chats/models to decrease RAM load across the board
* fix bug in thread safety for mpt model and de-duplicated code
* uses compact json format for network
* add remove model option in download dialog
",
    "contributors":
"
* Adam Treat (Nomic AI)
* Aaron Miller
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.4.3",
    "notes":
"
* add webserver feature that offers mirror api to chatgpt on localhost:4891
* add chatgpt models installed using openai key to chat client gui
* fixup the memory handling when switching between chats/models to decrease RAM load across the board
* fix bug in thread safety for mpt model and de-duplicated code
* uses compact json format for network
* add remove model option in download dialog
* remove text-davinci-003 as it is not a chat model
* fix installers on mac and linux to include libllmodel versions
",
    "contributors":
"
* Adam Treat (Nomic AI)
* Aaron Miller
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.4.4",
    "notes":
"
* fix buffer overrun in backend
* bugfix for browse for model directory
* dedup of qml code
* revamp settings dialog UI
* add localdocs plugin (beta) feature allowing scanning of local docs
* various other bugfixes and performance improvements
",
    "contributors":
"
* Adam Treat (Nomic AI)
* Aaron Miller
* Juuso Alasuutari
* Justin Wang
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.4.5",
    "notes":
"
* bugfix for model download remove
* bugfix for blocking on regenerate
* lots of various ui improvements enhancements
* big new change that brings us up2date with llama.cpp/ggml support for latest models
* advanced avx detection allowing us to fold the two installers into one
* new logging mechanism that allows for bug reports to have more detail
* make localdocs work with server mode
* localdocs fix for stale references after we regenerate
* fix so that browse to dialog on linux
* fix so that you can also just add a path to the textfield
* bugfix for chatgpt and resetting context
* move models.json to github repo so people can pr suggested new models
* allow for new models to be directly downloaded from huggingface in said prs
* better ui for localdocs settings
* better error handling when model fails to load
",
    "contributors":
"
* Nils Sauer (Nomic AI)
* Adam Treat (Nomic AI)
* Aaron Miller (Nomic AI)
* Richard Guo (Nomic AI)
* Konstantin Gukov
* Joseph Mearman
* Nandakumar
* Chase McDougall
* mvenditto
* Andriy Mulyar (Nomic AI)
* FoivosC
* Ettore Di Giacinto
* Tim Miller
* Peter Gagarinov
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.4.6",
    "notes":
"
* bugfix for model download remove
* bugfix for blocking on regenerate
* lots of various ui improvements enhancements
* big new change that brings us up2date with llama.cpp/ggml support for latest models
* advanced avx detection allowing us to fold the two installers into one
* new logging mechanism that allows for bug reports to have more detail
* make localdocs work with server mode
* localdocs fix for stale references after we regenerate
* fix so that browse to dialog on linux
* fix so that you can also just add a path to the textfield
* bugfix for chatgpt and resetting context
* move models.json to github repo so people can pr suggested new models
* allow for new models to be directly downloaded from huggingface in said prs
* better ui for localdocs settings
* better error handling when model fails to load
",
    "contributors":
"
* Nils Sauer (Nomic AI)
* Adam Treat (Nomic AI)
* Aaron Miller (Nomic AI)
* Richard Guo (Nomic AI)
* Konstantin Gukov
* Joseph Mearman
* Nandakumar
* Chase McDougall
* mvenditto
* Andriy Mulyar (Nomic AI)
* FoivosC
* Ettore Di Giacinto
* Tim Miller
* Peter Gagarinov
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.4.7",
    "notes":
"
* replit model support
* macos metal accelerated support
* fix markdown for localdocs references
* inline syntax highlighting for python and cpp with more languages coming
* synced with upstream llama.cpp
* ui fixes and default generation settings changes
* backend bugfixes
* allow for loading files directly from huggingface via TheBloke without name changes
",
    "contributors":
"
* Nils Sauer (Nomic AI)
* Adam Treat (Nomic AI)
* Aaron Miller (Nomic AI)
* Richard Guo (Nomic AI)
* Andriy Mulyar (Nomic AI)
* Ettore Di Giacinto
* AMOGUS
* Felix Zaslavskiy
* Tim Miller
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.4.8",
    "notes":
"
* replit model support
* macos metal accelerated support
* fix markdown for localdocs references
* inline syntax highlighting for python and cpp with more languages coming
* synced with upstream llama.cpp
* ui fixes and default generation settings changes
* backend bugfixes
* allow for loading files directly from huggingface via TheBloke without name changes
",
    "contributors":
"
* Nils Sauer (Nomic AI)
* Adam Treat (Nomic AI)
* Aaron Miller (Nomic AI)
* Richard Guo (Nomic AI)
* Andriy Mulyar (Nomic AI)
* Ettore Di Giacinto
* AMOGUS
* Felix Zaslavskiy
* Tim Miller
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.4.9",
    "notes":
"
* New GPT4All Falcon model
* New Orca models
* Token generation speed is now reported in GUI
* Bugfix for localdocs references when regenerating
* General fixes for thread safety
* Many fixes to UI to add descriptions for error conditions
* Fixes for saving/reloading chats
* Complete refactor of the model download dialog with metadata about models available
* Resume downloads bugfix
* CORS fix
* Documentation fixes and typos
* Latest llama.cpp update
* Update of replit
* Force metal setting
* Fixes for model loading with metal on macOS
",
    "contributors":
"
* Nils Sauer (Nomic AI)
* Adam Treat (Nomic AI)
* Aaron Miller (Nomic AI)
* Richard Guo (Nomic AI)
* Andriy Mulyar (Nomic AI)
* cosmic-snow
* AMOGUS
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.4.10",
    "notes":
"
* New GPT4All Falcon model
* New Orca models
* Token generation speed is now reported in GUI
* Bugfix for localdocs references when regenerating
* General fixes for thread safety
* Many fixes to UI to add descriptions for error conditions
* Fixes for saving/reloading chats
* Complete refactor of the model download dialog with metadata about models available
* Resume downloads bugfix
* CORS fix
* Documentation fixes and typos
* Latest llama.cpp update
* Update of replit
* Force metal setting
* Fixes for model loading with metal on macOS
",
    "contributors":
"
* Nils Sauer (Nomic AI)
* Adam Treat (Nomic AI)
* Aaron Miller (Nomic AI)
* Richard Guo (Nomic AI)
* Andriy Mulyar (Nomic AI)
* cosmic-snow
* AMOGUS
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.4.11",
    "notes":
"
* Per model settings
* Character settings
* Adding a system prompt
* Important bugfix for chatgpt install
* Complete refactor and revamp of settings dialog
* New syntax highlighting for java, bash, go
* Use monospace font for syntax highlighting of codeblocks
* New setting for turning off references in localdocs
* Fix memory leaks in falcon model
* Fix for backend memory handling
* Server mode bugfix
* Models.json retrieve bugfix
* Free metal context bugfix
* Add a close dialog feature to all chat dialogs
",
    "contributors":
"
* Lakshay Kansal (Nomic AI)
* Matthew Gill
* Brandon Beiler
* cosmic-snow
* Felix Zaslavskiy
* Andriy Mulyar (Nomic AI)
* Aaron Miller (Nomic AI)
* Adam Treat (Nomic AI)
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.4.12",
    "notes":
"
* Fix bad bug that was breaking numerous current installs (sorry folks!)
* Fix bug with 'browse' button in settings dialog
* Wayland support on linux
* Reduce template ui size in settings dialog
",
    "contributors":
"
* Akarshan Biswas
* Adam Treat (Nomic AI)
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.4.13",
    "notes":
"
* Fix bug with prolonging shutdown with generation
* Fix bug with update model info on deleting chats
* Fix bug with preventing closing of model download dialog
* Allows allow closing the model download dialog
* Fix numerous bugs with download of models.json and provide backup option
* Add json and c# highlighting
* Fix bug with chatgpt crashing
* Fix bug with chatgpt not working for some keys
* Fix bug with mixpanel opt outs not counting
* Fix problem with OOM errors causing crash and then repeating on next start
* Fix default thread setting and provide guardrails
* Fix tap handler in settings dialog for buttons
* Fix color of some text fields on macOS for settings dialog
* Fix problem with startup dialog not closing
* Provide error dialog for settings file not accessible
* Try and fix problems with avx-only detection
* Fix showing error in model downloads unnecessarily
* Prefer 7b models to load by default
* Add Wizard v1.1 to download list
* Rename Orca models to Mini Orca
* Don't use a system prompt unless model was trained with one by default
",
    "contributors":
"
* Lakshay Kansal (Nomic AI)
* Aaron Miller (Nomic AI)
* Adam Treat (Nomic AI)
* Community (beta testers, bug reporters)
"
  },
  {
    "version": "2.4.14",
    "notes":
"
* Add starcoder model support
* Add ability to switch between light mode/dark mode
* Increase the size of fonts in monospace code blocks a bit
",
    "contributors":
"
* Lakshay Kansal (Nomic AI)
* Adam Treat (Nomic AI)
"
  },
  {
    "version": "2.4.15",
    "notes":
"
* Add Vulkan GPU backend which allows inference on AMD, Intel and NVIDIA GPUs
* Add ability to switch font sizes
* Various bug fixes
",
    "contributors":
"
* Adam Treat (Nomic AI)
* Aaron Miller (Nomic AI)
* Nils Sauer (Nomic AI)
* Lakshay Kansal (Nomic AI)
"
  },
  {
    "version": "2.4.16",
    "notes":
"
* Bugfix for properly falling back to CPU when GPU can't be used
* Report the actual device we're using
* Fix context bugs for GPU accelerated models
",
    "contributors":
"
* Adam Treat (Nomic AI)
* Aaron Miller (Nomic AI)
"
  },
  {
    "version": "2.4.17",
    "notes":
"
* Bugfix for properly falling back to CPU when GPU is out of memory
",
    "contributors":
"
* Adam Treat (Nomic AI)
* Aaron Miller (Nomic AI)
"
  },
  {
    "version": "2.4.18",
    "notes":
"
* Bugfix for devices to show up in the settings combobox on application start and not just on model load
* Send information on requested device and actual device on model load to help assess which model/gpu/os combos are working
",
    "contributors":
"
* Adam Treat (Nomic AI)
"
  },
  {
    "version": "2.4.19",
    "notes":
"
* Fix a crash on systems with corrupted vulkan drivers or corrupted vulkan dlls
",
    "contributors":
"
* Adam Treat (Nomic AI)
"
  },
  {
    "version": "2.5.0",
    "notes":
"
* Major new release supports GGUF models only!
* New models like Mistral Instruct, Replit 1.5, Rift Coder and more
* All previous version of ggml-based models are no longer supported
* Extensive changes to vulkan support
* Better GPU error messages
* Prompt processing on the GPU
* Save chats now saves to text (less harddrive space)
* Many more changes
",
    "contributors":
"
* Aaron Miller (Nomic AI)
* Jared Van Bortel (Nomic AI)
* Adam Treat (Nomic AI)
* Community (beta testers, bug reporters, bindings authors)
"
  },
  {
    "version": "2.5.1",
    "notes":
"
* Accessibility fixes
* Bugfix for crasher on Windows
",
    "contributors":
"
* Aaron Miller (Nomic AI)
* Jared Van Bortel (Nomic AI)
* Victor Tsaran <vtsaran@yahoo.com>
* Community (beta testers, bug reporters, bindings authors)
"
  },
  {
    "version": "2.5.2",
    "notes":
"
* Support for GGUF v3 models
* Important fixes for AMD GPUs
* Don't start recalculating context immediately for saved chats
* UI fixes for chat name generation
* UI fixes for leading whitespaces in chat generation
",
    "contributors":
"
* Jared Van Bortel (Nomic AI)
* Adam Treat (Nomic AI)
* Community (beta testers, bug reporters, bindings authors)
"
  },
  {
    "version": "2.5.3",
    "notes":
"
* Major feature update for localdocs!
* Localdocs now uses an embedding model for retrieval augmented generation
* Localdocs can now search while your collections are indexing
* You're guaranteed to get hits from localdocs for every prompt you enter
* Fix: AMD gpu fixes
* Fix: Better error messages
",
    "contributors":
"
* Jared Van Bortel (Nomic AI)
* Adam Treat (Nomic AI)
* Community (beta testers, bug reporters, bindings authors)
"
  },
  {
    "version": "2.5.4",
    "notes":
"
* Major bugfix release with new models!
* Model: Recently released Orca 2 model which does exceptionally well on reasoning tasks
* Fix: System prompt was not always being honored
* Fix: Download network retry on cloudflare errors
",
    "contributors":
"
* Adam Treat (Nomic AI)
* Jared Van Bortel (Nomic AI)
* Community (beta testers, bug reporters, bindings authors)
"
  },
  {
    "version": "2.6.1",
    "notes":
"
* Update to newer llama.cpp
* Implemented configurable context length
* Bugfixes for localdocs
* Bugfixes for serialization to disk
* Bugfixes for AVX
* Bugfixes for Windows builds
* Bugfixes for context retention and clearing
* Add a button to collections dialog
",
    "contributors":
"
* Jared Van Bortel (Nomic AI)
* Adam Treat (Nomic AI)
* Community (beta testers, bug reporters, bindings authors)
"
  },
  {
    "version": "2.6.2",
    "notes":
"
* Update to latest llama.cpp
* Update to newly merged vulkan backend
* Partial GPU offloading support
* New localdocs speed increases and features
* New GUI settings option for configuring how many layers to put on GPU
* New lightmode theme, darkmode theme and legacy theme
* Lots of UI updates and enhancements
* Scores of bugfixes for stability and usability
",
    "contributors":
"
* Jared Van Bortel (Nomic AI)
* Adam Treat (Nomic AI)
* Karthik Nair
* Community (beta testers, bug reporters, bindings authors)
"
  },
  {
    "version": "2.7.0",
    "notes":
"
* Add support for twelve new model architectures
* Including Baichuan, BLOOM, CodeShell, GPT-2, Orion, Persimmon, Phi and Phi-2, Plamo, Qwen, Qwen2, Refact, and StableLM
* Fix for progress bar colors on legacy theme
* Fix sizing for model download dialog elements
* Fix dialog sizes to use more screen realestate where available
* Fix for vram leak when model loading fails
* Fix for making the collection dialog progress bar more readable
* Fix for smaller minimum size for main screen
* Fix for mistral crash
* Fix for mistral openorca prompt template to ChatLM
* Fix for excluding non-text documents from localdoc scanning
* Fix for scrollbar missing on main conversation
* Fix accessibility issues for screen readers
* Fix for not showing the download button when not online
",
    "contributors":
"
* Jared Van Bortel (Nomic AI)
* Adam Treat (Nomic AI)
* Community (beta testers, bug reporters, bindings authors)
"
  },
  {
    "version": "2.7.1",
    "notes":
"
* Update to latest llama.cpp with support for Google Gemma
* Gemma, Phi and Phi-2, Qwen2, and StableLM are now all GPU accelerated
* Large revamp of the model loading to support explicit unload/reload
* Bugfixes for ChatML and improved version of Mistral OpenOrca
* We no longer load a model by default on application start
* We no longer load a model by default on chat context switch
* Fixes for visual artifacts in update reminder dialog
* Blacklist Intel GPU's for now as we don't support yet
* Fixes for binary save/restore of chat
* Save and restore of window geometry across application starts
",
    "contributors":
"
* Jared Van Bortel (Nomic AI)
* Adam Treat (Nomic AI)
* Community (beta testers, bug reporters, bindings authors)
"
  },
  {
    "version": "2.7.2",
    "notes":
"
* New support for model search/discovery using huggingface search in downloads
* Support for more model architectures for GPU acceleration
* Three different crash fixes for corner case settings
* Add a minp sampling parameter
* Bert layoer norm epsilon value
* Fix problem with blank lines between reply and next prompt
",
    "contributors":
"
* Christopher Barrera
* Jared Van Bortel (Nomic AI)
* Adam Treat (Nomic AI)
* Community (beta testers, bug reporters, bindings authors)
"
  },
  {
    "version": "2.7.3",
    "notes":
"
* Fix for network reachability unknown
* Fix undefined behavior with resetContext
* Fix ChatGPT which was broken with previous release
* Fix for clean up of chat llm thread destruction
* Display of model loading warnings
* Fix for issue 2080 where the GUI appears to hang when a chat is deleted
* Fix for issue 2077 better responsiveness of model download dialog when download is taking place
* Fix for issue 2092 don't include models that are disabled for GUI in application default model list
* Fix for issue 2087 where cloned modelds were lost and listed in download dialog erroneously
* Fix for MPT models without duplicated token embd weight
* New feature with api server port setting
* Fix for issue 2024 where combobox for model settings uses currently used model by default
* Clean up settings properly for removed models and don't list stale model settings in download dialog
* Fix for issue 2105 where the cancel button was not working for discovered model downloads
",
    "contributors":
"
* Christopher Barrera
* Daniel Alencar
* Jared Van Bortel (Nomic AI)
* Adam Treat (Nomic AI)
* Community (beta testers, bug reporters, bindings authors)
"
  },
  {
    "version": "2.7.4",
    "notes":
"
<b>&mdash; What's New &mdash;</b>
* Add a right-click menu to the chat (by @kryotek777 in PR #2108)
* Change the left sidebar to stay open (PR #2117)
* Limit the width of text in the chat (PR #2118)
* Move to llama.cpp's SBert implementation (PR #2086)
* Support models provided by the Mistral AI API (by @Olyxz16 in PR #2053)
* Models List: Add Ghost 7B v0.9.1 (by @lh0x00 in PR #2127)
* Add Documentation and FAQ links to the New Chat page (by @3Simplex in PR #2183)
* Models List: Simplify Mistral OpenOrca system prompt (PR #2220)
* Models List: Add Llama 3 Instruct (PR #2242)
* Models List: Add Phi-3 Mini Instruct (PR #2252)
* Improve accuracy of anonymous usage statistics (PR #2238)

<b>&mdash; Fixes &mdash;</b>
* Detect unsupported CPUs correctly on Windows (PR #2141)
* Fix the colors used by the server chat (PR #2150)
* Fix startup issues when encountering non-Latin characters in paths (PR #2162)
* Fix issues causing LocalDocs context links to not work sometimes (PR #2218)
* Fix incorrect display of certain code block syntax in the chat (PR #2232)
* Fix an issue causing unnecessary indexing of document collections on startup (PR #2236)
",
    "contributors":
"
* Jared Van Bortel (Nomic AI)
* Adam Treat (Nomic AI)
* Lam Hieu (`@lh0x00`)
* 3Simplex (`@3Simplex`)
* Kryotek (`@kryotek777`)
* Olyxz16 (`@Olyxz16`)
* Robin Verduijn (`@robinverduijn`)
* Tim453 (`@Tim453`)
* Xu Zhen (`@xuzhen`)
* Community (beta testers, bug reporters, bindings authors)
"
  },
  {
    "version": "2.7.5",
    "notes":
"
<b>&mdash; What's New &mdash;</b>
* Improve accuracy of anonymous usage statistics (PR #2297, PR #2299)

<b>&mdash; Fixes &mdash;</b>
* Fix some issues with anonymous usage statistics (PR #2270, PR #2296)
* Default to GPU with most VRAM on Windows and Linux, not least (PR #2297)
* Fix initial failure to generate embeddings with Nomic Embed (PR #2284)
",
    "contributors":
"
* Jared Van Bortel (Nomic AI)
* Adam Treat (Nomic AI)
* Community (beta testers, bug reporters, bindings authors)
"
  },
  {
    "version": "2.8.0",
    "notes":
"
<b>&mdash; What's New &mdash;</b>
* Context Menu: Replace \"Select All\" on message with \"Copy Message\" (PR #2324)
* Context Menu: Hide Copy/Cut when nothing is selected (PR #2324)
* Improve speed of context switch after quickly switching between several chats (PR #2343)
* New Chat: Always switch to the new chat when the button is clicked (PR #2330)
* New Chat: Always scroll to the top of the list when the button is clicked (PR #2330)
* Update to latest llama.cpp as of May 9, 2024 (PR #2310)
* **Add support for the llama.cpp CUDA backend** (PR #2310, PR #2357)
  * Nomic Vulkan is still used by default, but CUDA devices can now be selected in Settings
  * When in use: Greatly improved prompt processing and generation speed on some devices
  * When in use: GPU support for Q5\_0, Q5\_1, Q8\_0, K-quants, I-quants, and Mixtral
* Add support for InternLM models (PR #2310)

<b>&mdash; Fixes &mdash;</b>
* Do not allow sending a message while the LLM is responding (PR #2323)
* Fix poor quality of generated chat titles with many models (PR #2322)
* Set the window icon correctly on Windows (PR #2321)
* Fix a few memory leaks (PR #2328, PR #2348, PR #2310)
* Do not crash if a model file has no architecture key (PR #2346)
* Fix several instances of model loading progress displaying incorrectly (PR #2337, PR #2343)
* New Chat: Fix the new chat being scrolled above the top of the list on startup (PR #2330)
* macOS: Show a \"Metal\" device option, and actually use the CPU when \"CPU\" is selected (PR #2310)
* Remove unsupported Mamba, Persimmon, and PLaMo models from the whitelist (PR #2310)
* Fix GPT4All.desktop being created by offline installers on macOS (PR #2361)
",
    "contributors":
"
* Jared Van Bortel (Nomic AI)
* Adam Treat (Nomic AI)
* Tim453 (`@Tim453`)
* Community (beta testers, bug reporters, bindings authors)
"
  },
  {
    "version": "3.0.0",
    "notes":
"
<b>&mdash; What's New &mdash;</b>
* Complete UI overhaul (PR #2396)
* LocalDocs improvements (PR #2396)
  * Use nomic-embed-text-v1.5 as local model instead of SBert
  * Ship local model with application instead of downloading afterwards
  * Store embeddings flat in SQLite DB instead of in hnswlib index
  * Do exact KNN search with usearch instead of approximate KNN search with hnswlib
* Markdown support (PR #2476)
* Support CUDA/Metal device option for embeddings (PR #2477)

<b>&mdash; Fixes &mdash;</b>
* Fix embedding tokenization after PR #2310 (PR #2381)
* Fix a crash when loading certain models with \"code\" in their name (PR #2382)
* Fix an embedding crash with large chunk sizes after PR #2310 (PR #2383)
* Fix inability to load models with non-ASCII path on Windows (PR #2388)
* CUDA: Do not show non-fatal DLL errors on Windows (PR #2389)
* LocalDocs fixes (PR #2396)
  * Always use requested number of snippets even if there are better matches in unselected collections
  * Check for deleted files on startup
* CUDA: Fix PTX errors with some GPT4All builds (PR #2421)
* Fix blank device in UI after model switch and improve usage stats (PR #2409)
* Use CPU instead of CUDA backend when GPU loading fails the first time (ngl=0 is not enough) (PR #2477)
* Fix crash when sending a message greater than n\_ctx tokens after PR #1970 (PR #2498)
",
    "contributors":
"
* Vincent Giardina (Nomic AI)
* Jared Van Bortel (Nomic AI)
* John W. Parent (Kitware)
* Paige Lee (Nomic AI)
* Max Cembalest (Nomic AI)
* Andriy Mulyar (Nomic AI)
* Adam Treat (Nomic AI)
* cosmic-snow (`@cosmic-snow`)
* Community (beta testers, bug reporters, bindings authors)
"
  },
  {
    "version": "3.1.0",
    "notes":
"
<b>&mdash; What's New &mdash;</b>
* Generate suggested follow-up questions feature (#2634)

<b>&mdash; What's Changed &mdash;</b>
* Customize combo boxes and context menus to fit the new style (#2535)
* Improve view bar scaling and Model Settings layout (#2520)
* Make the logo spin while the model is generating (#2557)
* Server: Reply to wrong GET/POST method with HTTP 405 instead of 404 (#2615)
* Update theme for menus (#2578)
* Move the \"stop\" button to the message box (#2561)
* Build with CUDA 11.8 for better compatibility (#2639)
* Make links in latest news section clickable (#2643)
* Support translation of settings choices (#2667), (#2690)
* Improve LocalDocs view's error message (by @cosmic-snow in #2679)
* Ignore case of LocalDocs file extensions (#2642), (#2684)
* Update llama.cpp to commit 87e397d00 from July 19th (#2694)
  * Add support for GPT-NeoX, Gemma 2, OpenELM, ChatGLM, and Jais architectures (all with Vulkan support)
  * Enable Vulkan support for StarCoder2, XVERSE, Command R, and OLMo
* Show scrollbar in chat collections list as needed (#2691)

<b>&mdash; What's Removed &mdash;</b>
* Remove support for GPT-J models (#2676)

<b>&mdash; Fixes &mdash;</b>
* Fix placement of thumbs-down and datalake opt-in dialogs (#2540)
* Select the correct folder with the Linux fallback folder dialog (#2541)
* Fix clone button sometimes producing blank model info (#2545)
* Fix jerky chat view scrolling (#2555)
* Fix \"reload\" showing for chats with missing models (#2520)
* Fix property binding loop warning (#2601)
* Fix UI hang with certain chat view content (#2543)
* Fix crash when Kompute falls back to CPU (#2640)
* Fix several Vulkan resource management issues (#2694)
",
    "contributors":
"
* Jared Van Bortel (Nomic AI)
* Adam Treat (Nomic AI)
* cosmic-snow (`@cosmic-snow`)
* 3Simplex (`@3Simplex`)
* Community (beta testers, bug reporters, bindings authors)
"
  }
]