Commit Graph

49 Commits (02c9bb4ac72b7fcb21a9bd36a7c05688531ec848)

Author SHA1 Message Date
Adam Treat c3d923cdc7 Don't add new chats willy nilly. 1 year ago
Adam Treat d91dd567e2 Hot swapping of conversations. Destroys context for now. 1 year ago
Adam Treat 925ad70833 Turn the chat list into a model. 1 year ago
Adam Treat 463c1474dc Provide convenience methods for adding/removing/changing chat. 1 year ago
Adam Treat 482f543675 Handle the fwd of important signals from LLM object so qml doesn't have to deal with which chat is current. 1 year ago
Adam Treat 414a12c33d Major refactor in prep for multiple conversations. 1 year ago
Adam Treat bbffa7364b Add new C++ version of the chat model. Getting ready for chat history. 1 year ago
Adam Treat 037a9a6ec5 Remove these as it is mitigated by repeat penalty and models really should train this out. 1 year ago
Adam Treat 9b467f2dee Use the universal sep. 1 year ago
Adam Treat 2a5b34b193 Load models from filepath only. 1 year ago
Adam Treat c6c5e0bb4f Always try and load default model first. Groovy is the default default. 1 year ago
Adam Treat a3253c4ab1 Move the saving of the tokens to the impl and not the callbacks responsibility. 1 year ago
Adam Treat 9a65f73392 Move the promptCallback to own function. 1 year ago
Adam Treat 5c3c1317f8 Track check for updates. 1 year ago
Adam Treat eafb98b3a9 Initial support for opt-in telemetry. 1 year ago
Adam Treat 70e6b45123 Don't crash when prompt is too large. 1 year ago
Aaron Miller aa20bafc91 new settings (model path, repeat penalty) w/ tabs 1 year ago
Adam Treat b6937c39db Infinite context window through trimming. 1 year ago
Adam Treat 8b1ddabe3e Implement repeat penalty for both llama and gptj in gui. 1 year ago
Adam Treat cd2e559db4 Don't crash right out of the installer ;) 1 year ago
Aaron Miller 6e92d93b53 persistent threadcount setting
threadcount is now on the Settings object and
gets reapplied after a model switch
1 year ago
Adam Treat e4b110639c Add a fixme for dubious code. 1 year ago
Adam Treat 29685b3eab Provide a non-priviledged place for model downloads when exe is installed to root. 1 year ago
Adam Treat 795715fb59 Don't crash starting with no model. 1 year ago
Adam Treat 71b308e914 Add llama.cpp support for loading llama based models in the gui. We now
support loading both gptj derived models and llama derived models.
1 year ago
Aaron Miller 00cb5fe2a5 Add thread count setting 1 year ago
Adam Treat 169afbdc80 Add a new model download feature. 1 year ago
Adam Treat 2b1cae5a7e Allow unloading/loading/changing of models. 1 year ago
Adam Treat f73fbf28a4 Fix the context. 1 year ago
Adam Treat a7c2d65824 Don't allow empty prompts. Context past always equal or greater than zero. 1 year ago
Adam Treat 4bf4b2a080 Trim trailing whitespace at the end of generation. 1 year ago
Adam Treat 9381a69b2b Remove newlines too. 1 year ago
Adam Treat b39acea516 More conservative default params and trim leading whitespace from response. 1 year ago
Aaron Miller 5bfb3f8229 use the settings dialog settings when generating 1 year ago
Adam Treat a77946e745 Provide an instruct/chat template. 1 year ago
Aaron Miller 391904efae Use completeBaseName to display model name
this cuts the filename at the *final* dot instead of the first, allowing
model names with version numbers to be displayed correctly.
1 year ago
Adam Treat 078b755ab8 Erase the correct amount of logits when regenerating which is not the same
as the number of tokens.
1 year ago
Adam Treat b1bb9866ab Fix crash with recent change to erase context. 1 year ago
Adam Treat 1c5dd6710d When regenerating erase the previous response and prompt from the context. 1 year ago
Adam Treat a9eced2d1e Add an abstraction around gpt-j that will allow other arch models to be loaded in ui. 1 year ago
Adam Treat 661191ce12 Fix the check for updates on mac. 1 year ago
Adam Treat a06fd8a487 Provide a busy indicator if we're processing a long prompt and make the
stop button work in the middle of processing a long prompt as well.
1 year ago
Adam Treat 1e13f8648c Fix the name of the updates tool. 1 year ago
Adam Treat 01dee6f20d Programmatically get the model name from the LLM. The LLM now searches
for applicable models in the directory of the executable given a pattern
match and then loads the first one it finds.

Also, add a busy indicator for model loading.
1 year ago
Adam Treat f1bbe97a5c Big updates to the UI. 1 year ago
Adam Treat c62ebdb81c Add a reset context feature to clear the chat history and the context for now. 1 year ago
Adam Treat b088929df4 Fixes for linux and macosx. 1 year ago
Adam Treat c763d4737d Prelim support for past context. 1 year ago
Adam Treat ff2fdecce1 Initial commit. 1 year ago