Jared Van Bortel

### Feature Request Several times I have heard that users have found system prompts that improve the accuracy of LocalDocs retrieval. e.g., https://github.com/nomic-ai/gpt4all/discussions/1766#discussioncomment-8771690 Using a given system prompt across multiple...

[Feature] Ability to check for updates to discovered models

2

### Discussed in https://github.com/nomic-ai/gpt4all/discussions/2064 Originally posted by **SINAPSA-IC** March 2, 2024 Hello. Assuming that there's an update for an LLM that we use in GPT4All, how can we know this?...

GPT4All does not detect my GPU

1

### Discussed in https://github.com/nomic-ai/gpt4all/discussions/2115 Originally posted by **TerrificTerry** March 13, 2024 I'm currently trying out the Mistra OpenOrca model, but it only runs on CPU with 6-7 tokens/sec. My laptop...

Expose llama.cpp's progress_callback to bindings

2

We could expose llama.cpp's progress_callback to provide a way to both report progress and cancel model loading via the bindings. ref #1934

Executable bit set on non-executable files

1

A seemingly random set of files (including many JSON files and C source files) are set executable. I run Linux and use zsh, but given an environment with a fairly...

Doesn't support relaxed root value restrictions in RFC 8259

1

All tests from [nst/JSONTestSuite](https://github.com/nst/JSONTestSuite) with a single value (string, number, boolean, or null) at the root fail with an error: ``` document root must be object or array ``` One...

Fixes "WARNING: Request to generate sync embeddings for non-local model invalid" when using Nomic Embed.

Type of user_data is None throughout the stubs

1

user_data is obviously meant for the programmer to put arbitrary data in for later use. At runtime it may be any type. However, user_data is typed as None throughout these...

Support GLib.option (or use a better placeholder)

2

Here is a simple example: ```python from gi.repository import GLib parser = GLib.option.OptionParser(option_list=[ GLib.option.make_option('--flag', action='store_true', help='flag'), ]) parser.parse_args() print('flag is', parser.values.flag) ``` mypy complains about type errors: ``` $ mypy...

CUDA backend

5

This PR adds opt-in CUDA support in the GPT4All UI and python bindings using the llama.cpp CUDA backend. CUDA-enabled devices will appear as e.g. "CUDA: Tesla P40" on supported platforms,...