GPT4All: vanhemmat versiot ja versiohistoria

What's New
- Remote Models:
- The Add Model page now has a dedicated tab for remote model providers.
- Groq, OpenAI, and Mistral remote models are now easier to configure.
- CUDA Compatibility: GPUs with CUDA compute capability 5.0 such as the GTX 750 are now supported by the CUDA backend.
- New Model: The non-MoE Granite model is now supported.
- Translation Updates:
- The Italian translation has been updated.
- The Simplified Chinese translation has been significantly improved.
- Better Chat Templates: The default chat templates for OLMoE 7B 0924/0125 and Granite 3.1 3B/8B have been improved.
- Whitespace Fixes: DeepSeek-R1-based models now have better whitespace behavior in their output.
- Crash Fixes: Several issues that could potentially cause GPT4All to crash have been fixed.
Contributors
- Jared Van Bortel (Nomic AI)
- Adam Treat (Nomic AI)
- ThiloteE (@ThiloteE)
- Lil Bob (@Junior2Ran)
- Riccardo Giovanetti (@Harvester62)
Full Changelog: CHANGELOG.md

What's New
- LocalDocs Fix: LocalDocs no longer shows an error on later messages with reasoning models.
- DeepSeek Fix: DeepSeek-R1 reasoning (in 'think' tags) no longer appears in chat names and follow-up questions.
- Windows ARM Improvements:
- Graphical artifacts on some SoCs have been fixed.
- A crash when adding a collection of PDFs to LocalDocs has been fixed.
- Template Parser Fixes: Chat templates containing an unclosed comment no longer freeze GPT4All.
- New Models: OLMoE and Granite MoE models are now supported.
Contributors
- Jared Van Bortel (Nomic AI)
- Adam Treat (Nomic AI)
- ThiloteE (@ThiloteE)
Full Changelog: CHANGELOG.md

What's New
- Native DeepSeek-R1-Distill Support: GPT4All now has robust support for the DeepSeek-R1 family of distillations.
- Several model variants are now available on the downloads page.
- Reasoning (wrapped in "think" tags) is displayed similarly to the Reasoner model.
- The DeepSeek-R1 Qwen pretokenizer is now supported, resolving the loading failure in previous versions.
- The model is now configured with a GPT4All-compatible prompt template by default.
- Chat Templating Overhaul: The template parser has been completely replaced with one that has much better compatibility with common models.
- Code Interpreter Fixes:
- An issue preventing the code interpreter from logging a single string in v3.7.0 has been fixed.
- The UI no longer freezes while the code interpreter is running a computation.
- Local Server Fixes:
- An issue preventing the server from using LocalDocs after the first request since v3.5.0 has been fixed.
- System messages are now correctly hidden from the message history.
Contributors
- Jared Van Bortel (Nomic AI)
- Adam Treat (Nomic AI)
- Riccardo Giovanetti (@Harvester62)
Full Changelog: CHANGELOG.md

What's New
- Windows ARM Support: GPT4All now supports the Windows ARM platform, ensuring compatibility with devices powered by Qualcomm Snapdragon and Microsoft SQ-series processors.
- NOTE: Support for GPU and/or NPU acceleration is not available at this time. Only the CPU will be used to run LLMs.
- NOTE: You must install the new Windows ARM version of GPT4All from the website. The standard Windows version will not work due to emulation limitations.
- Fixed Updating on macOS: The maintenance tool no longer crashes when attempting to update or uninstall GPT4All on Sequoia.
- NOTE: If you have installed the version from the GitHub releases as a workaround for this issue, you can safely uninstall it and switch back to the version from the website.
- Fixed Chat Saving on macOS: Chats now save as expected when the application is quit with Command-Q.
- Code Interpreter Improvements:
- The behavior when the code takes too long to execute and times out has been improved.
- console.log now accepts multiple arguments for better compatibility with native JavaScript.
- Chat Templating Improvements:
- Two crashes and one compatibility issue have been fixed in the chat template parser.
- The default chat template for EM German Mistral has been fixed.
- Automatic replacements have been added for five new models as we continue to improve compatibility with common chat templates.
Contributors
- Jared Van Bortel (Nomic AI)
- Adam Treat (Nomic AI)
- Riccardo Giovanetti (@Harvester62)
Full Changelog: CHANGELOG.mdNote: Please be careful when downloading an offline installer below. There is now a -win64-arm installer. If you have an ARM processor (e.g. Qualcomm Snapdragon or Microsoft SQ-series), use the -win64-arm one. If you have an Intel or AMD processor, or if unsure, use the -win64 one.

What's New
- Fixes:
- The stop generation button no longer working in v3.6.0.
- The copy entire conversation button no longer working in v3.6.0.
Contributors
- Adam Treat (Nomic AI)
Full Changelog: CHANGELOG.md

What's New
- LocalDocs Fixes: Several issues with LocalDocs, some of which were introduced in v3.4.0, have been fixed.
- Fixed the possible use of references from unselected collections.
- Fixed unnecessary reindexing of files with uppercase extensions.
- Fixed hybrid search failure due to inconsistent database state.
- Fully fixed the blank Embeddings Device selection in LocalDocs settings.
- Fixed LocalDocs indexing of large PDFs making very slow progress or even stalling.
Contributors
- Adam Treat (Nomic AI)
- Jared Van Bortel (Nomic AI)
Full Changelog: CHANGELOG.md

Added
- Generate suggested follow-up questions (#2634)
- Scaffolding for translations (#2612)
- Spanish (MX) translation (by @jstayco in #2654)
- Chinese (Simplified) translation by mikage (#2657)
- Dynamic changes of language and locale at runtime (#2659, #2677)
- Romanian translation by @SINAPSA_IC (#2662)
- Chinese (Traditional) translation (by @supersonictw in #2661)
- Italian translation (by @Harvester62 in #2700)
Changed
- 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 (by @cosmic-snow in #2615)
- Update theme for menus (by @3Simplex in #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 (by @cosmic-snow in #2691)
Removed
- Remove support for GPT-J models (#2676, #2693)
Fixed
- 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)

What's New
- Complete UI overhaul (#2396)
- LocalDocs improvements (#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 (#2476)
- Support CUDA/Metal device option for embeddings (#2477)
Fixes
- Fix embedding tokenization after #2310 (#2381)
- Fix a crash when loading certain models with "code" in their name (#2382)
- Fix an embedding crash with large chunk sizes after #2310 (#2383)
- Fix inability to load models with non-ASCII path on Windows (#2388)
- CUDA: Do not show non-fatal DLL errors on Windows (#2389)
- LocalDocs fixes (#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 (#2421)
- Fix blank device in UI after model switch and improve usage stats (#2409)
- Use CPU instead of CUDA backend when GPU loading fails the first time (ngl=0 is not enough) (#2477)
- Fix crash when sending a message greater than n_ctx tokens after #1970 (#2498)
New Contributors
- @woheller69 made their first contribution in (#2339)
- @patcher9 made their first contribution in (#2386)
- @sunsided made their first contribution in (#2414)
- @johnwparent made their first contribution in (#2319)
- @mcembalest made their first contribution in (#2488)
Full Changelog: https://github.com/nomic-ai/gpt4all/compare/v2.8.0...v3.0.0

What's New
- Context Menu: Replace "Select All" on message with "Copy Message" (#2324)
- Context Menu: Hide Copy/Cut when nothing is selected (#2324)
- Improve speed of context switch after quickly switching between several chats (#2343)
- New Chat: Always switch to the new chat when the button is clicked (#2330)
- New Chat: Always scroll to the top of the list when the button is clicked (#2330)
- Update to latest llama.cpp as of May 9, 2024 (#2310)
- Add support for the llama.cpp CUDA backend (#2310, #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 (#2310)
Fixes
- Do not allow sending a message while the LLM is responding (#2323)
- Fix poor quality of generated chat titles with many models (#2322)
- Set the window icon correctly on Windows (#2321)
- Fix a few memory leaks (#2328, #2348, #2310)
- Do not crash if a model file has no architecture key (#2346)
- Fix several instances of model loading progress displaying incorrectly (#2337, #2343)
- New Chat: Fix the new chat being scrolled above the top of the list on startup (#2330)
- macOS: Show a "Metal" device option, and actually use the CPU when "CPU" is selected (#2310)
- Remove unsupported Mamba, Persimmon, and PLaMo models from the whitelist (#2310)
- Fix GPT4All.desktop being created by offline installers on macOS (#2361)
Full Changelog: https://github.com/nomic-ai/gpt4all/compare/v2.7.5...v2.8.0

What's Changed
- Add 12 new model architectures for CPU and Metal inference (#1914)
- These are Baichuan, BLOOM, CodeShell, GPT-2, Orion, Persimmon, Phi and Phi-2, Plamo, Qwen, Qwen2, Refact, and StableLM.
- We don't have official downloads for these yet, but TheBloke offers plenty of compatible GGUF quantizations.
- Restore minimum window size of 720x480 (1b524c4)
- Use ChatML for Mistral OpenOrca to make its output formatting more reliable (#1935)
Bug Fixes
- Fix VRAM not being freed when CPU fallback occurs - this makes switching models more reliable (#1901)
- Disable offloading of Mixtral to GPU because we crash otherwise (#1931)
- Limit extensions scanned by LocalDocs to txt, pdf, md, rst - other formats were inserting useless binary data (#1912)
- Fix missing scrollbar for chat history (490404d)
- Accessibility improvements (4258bb1)
New Contributors
- @boshk0 made their first contribution in #1924