* rebase onto llama.cpp commit ggerganov/llama.cpp@d46dbc76f
* support for CUDA backend (enabled by default)
* partial support for Occam's Vulkan backend (disabled by default)
* partial support for HIP/ROCm backend (disabled by default)
* sync llama.cpp.cmake with upstream llama.cpp CMakeLists.txt
* changes to GPT4All backend, bindings, and chat UI to handle choice of llama.cpp backend (Kompute or CUDA)
* ship CUDA runtime with installed version
* make device selection in the UI on macOS actually do something
* model whitelist: remove dbrx, mamba, persimmon, plamo; add internlm and starcoder2
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
* select the existing new chat if there already is one when "New Chat" is clicked
* scroll to the new chat when "New Chat" is clicked
* fix the "New Chat" being scrolled past the top of the chat list
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
* prevent load progress from getting out of sync with the current chat
* fix memory leak on exit if the LLModelStore contains a model
* do not report cancellation as a failure in console/Mixpanel
* show "waiting for model" separately from "switching context" in UI
* do not show lower "reload" button on error
* skip context switch if unload is pending
* skip unnecessary calls to LLModel::saveState
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
* Fix destruction and tear down of the embedding thread.
Signed-off-by: Adam Treat <treat.adam@gmail.com>
* Fix order of deletion to prevent use after free.
Signed-off-by: Adam Treat <treat.adam@gmail.com>
---------
Signed-off-by: Adam Treat <treat.adam@gmail.com>
* chat: fix window icon on Windows
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
* chat: remove redundant copy of macOS app icon
This has been redundant since PR #2180.
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
---------
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
* ChatView: fix deprecation warning
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
* ChatView: make context menus more intuitive
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
---------
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
* localdocs: load model before checking what we loaded
Fixes "WARNING: Request to generate sync embeddings for non-local model
invalid"
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
* fix inverted assertion
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
---------
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
Other changes:
- Always display first start dialog if privacy options are unset (e.g. if the user closed GPT4All without selecting them)
- LocalDocs scanQueue is now always deferred
- Fix a potential crash in magic_match
- LocalDocs indexing is now started after the first start dialog is dismissed so usage stats are included
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
Other changes:
* fix 'requires' for models with %2 in template
* move Ghost 7B to the appropriate location in the file based on where it actually appears in the UI
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
* chat: revert PR #2187
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
* chat: revert PR #2148
This reverts commit f571e7e450.
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
---------
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
* chat: use .rmodel extension for Nomic Embed
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
* database: fix order of SQL arguments in updateDocument
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
---------
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
Directed the Documentation link specifically to the ChatUI documentation.
Co-authored-by: ThiloteE <73715071+ThiloteE@users.noreply.github.com>
Signed-off-by: 3Simplex <10260755+3Simplex@users.noreply.github.com>
Include links for Documentation and FAQ for new users on the "new chat view".
Co-authored-by: ThiloteE <73715071+ThiloteE@users.noreply.github.com>
Signed-off-by: 3Simplex <10260755+3Simplex@users.noreply.github.com>
* chat: fix non-AVX CPU detection on Windows
* bindings: throw exception instead of logging to console
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
This commit allow changing the install path during CMake configure step using the CMAKE_INSTALL_PREFIX variable. If the variable is not set, it still defaults to {CMAKE_BINARY_DIR}/install.
Signed-off-by: Tim453 <50015765+Tim453@users.noreply.github.com>
Key changes:
* honor empty system prompt argument
* current_chat_session is now read-only and defaults to None
* deprecate fallback prompt template for unknown models
* fix mistakes from #2086
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
Signed-off-by: Jared Van Bortel <jared@nomic.ai>
Signed-off-by: Cédric Sazos <cedric.sazos@tutanota.com>
Co-authored-by: Jared Van Bortel <jared@nomic.ai>
valid because we expect the url to contain the actual filename at the end.
This also allows huggingface to track the download as happening.
Signed-off-by: Adam Treat <treat.adam@gmail.com>
Adds Nous-Hermes-2-Mistral-7B-DPO.Q4_0.gguf, which is the new 7b flagship model of NousResearch.
**Original Model location:**
https://huggingface.co/NousResearch/Nous-Hermes-2-Mistral-7B-DPO-GGUF
**Model description:**
Nous Hermes 2 on Mistral 7B DPO is the new flagship 7B Hermes! This model was DPO'd from Teknium/OpenHermes-2.5-Mistral-7B and has improved across the board on all benchmarks tested - AGIEval, BigBench Reasoning, GPT4All, and TruthfulQA.
The model prior to DPO was trained on 1,000,000 instructions/chats of GPT-4 quality or better, primarily synthetic data as well as other high quality datasets, available from the repository teknium/OpenHermes-2.5.
**Original Dataset Location:**
https://huggingface.co/datasets/teknium/OpenHermes-2.5
**Dataset description:**
This is the dataset that made OpenHermes 2.5 and Nous Hermes 2 series of models.
The Open Hermes 2/2.5 and Nous Hermes 2 models have made significant advancements of SOTA LLM's over recent months, and are underpinned by this exact compilation and curation of many open source datasets and custom created synthetic datasets.
The Open Hermes 2.5 dataset is a continuation of the Open Hermes 1 dataset, at a much larger scale, much more diverse, and much higher quality compilation, reaching 1M, primarily synthetically generated instruction and chat samples.
Signed-off-by: ThiloteE <73715071+ThiloteE@users.noreply.github.com>
