Commit History
-
Bump version to svelte 4 (#488)
Unverified signatureFix linting & type checksf504c92
-
move some tasks to small model (#479)
Unverified signaturebf15962 -
Remove shared routes (#478)
Unverified signature* rm shared route * fix sharing of shared convsa799675
-
Store preprompt in conversation (#422) (#470)
Unverified signature* Store preprompt in conversation (#422) * display preprompt at the top * system prompt modal * tweaks --------- Co-authored-by: Victor Mustar <victor.mustar@gmail.com>af58e08
-
add manifest.json (#469)
Unverified signature98030ef
-
fix title summarization
4ad4e35 -
remove logs for parsing
140086b
-
Add a message wide copy button (#453)
Unverified signature* add-copytoclipboardbtn for the all message * fix padding * fix padding * Fix styling * Move before like and dislike button * position and spacing * mobile fix --------- Co-authored-by: Victor Mustar <victor.mustar@gmail.com>a7dc1aa
-
fix JSON.parse for summerize (#445)
Unverified signature* fix JSON.parse for summerize When serving with TGI, summerize calls this function and it errors with `SyntaxError: Unexpected token d in JSON at position 0` This PR fixes the problem and keeps existing behaviour. * fix types --------- Co-authored-by: Nathan Sarrazin <sarrazin.nathan@gmail.com>1e19fc8 -
fix types
0134fe1
-
error console instead of crashing
5b07906 -
Make embedding model settings more future-proof (#454)
MishigUnverified signature3acc11d -
Refactor summarization so it gets called from backend (#456)
Unverified signature* Refactor summarization * get rid of debug log * remove old todo9960338 -
Backend refactor for streaming endpoints (#444)
Unverified signature* wip: complete refactor of streaming backend * working refactoring * fix missing first token & perf regression in output quality * lint * Fix websearch loading from db * fix loading * fix invalidate * remove logs * fix SSR error * typo: paragraphs * fixed save on abort * lint * lint * remove debug log in console * lint for reale943a05 -
Revert "Revert "Update embedding model for WebSearch (#437)""
This reverts commit 1061bc237c37ada9537197b10aeee43e08fc8f4e.aa07e29 -
Revert "Improve error message when the .env MODELS is not well formatted (#451)"
This reverts commit 8ce8b634e9687f00ad31d572dc047113e5b9f0d0.7ddda31 -
Revert "Update embedding model for WebSearch (#437)"
This reverts commit f88542b1121903c1df34b20bd7b8213148e85ca0.1061bc2 -
Update embedding model for WebSearch (#437)
MishigUnverified signature* Use `gte-base` as the emebdding model * use `bge-small-en-v1.5` * Revert "use `bge-small-en-v1.5`" This reverts commit 8cfe084d26884c130d6034afa095b7bbf0f7fa1b. * Use `gte-small`f88542b -
Improve error message when the .env MODELS is not well formatted (#451)
Unverified signature* feat: Improve error handling and parsing of MODELS environment variable * Add more verbose parsing error * Lint * improve message * lint * refactor error handling and default values in models * improve * format --------- Co-authored-by: Nathan Sarrazin <sarrazin.nathan@gmail.com>8ce8b63 -
Add latex support with marked-katex-extension (#450)
Unverified signature* Add latex support with marked-katex-extension * Add renderer * Fix marked default option problem * Fix linting error * Fix lock error15bf16f -
fix lint on main
77df078 -
Mobile: fix model selection (#448)
Unverified signature* adjustments and mobile modal * use dvh unit * marginc867764
-
[Websearch] update (#427)
Mishig +2: julien-c, victorUnverified signature* Fix reuqest body * update webSearchQueryPromptTemplate * update generate google query parser * Add today's date to google search query creator * crawl top stories if exts; remove answer_box & knowledgeGraph * Create paragraph chunks from top articles * flattened paragprah chunks * update status texts * add gradio client * call gradio app for RAG * Web scrape only "p, li, span" els * add MAX_N_CHUNKS * gradio result typing * parse only <p> elements * rm dev change * update typing WebSearch * buld RAG prompt * Rm dev change * change websearch context msg from user to assisntat type * use hosted gradio app * fix lint * prompt engineering * more prompt engineering * MAX_N_PAGES_SCRAPE = 10 * better error msg * more prompt engineering * revert websearch prompt to previous * rm `top_stories` from websearch as the results are not good * Stop using gradio client, use regular fetch * chore * Rm websearch…ebac87f -
Add rate limiting to websearch and title summary (#433)
Unverified signature0953d85
-
move rate limit event to top
ba93cf8 -
ip rate limit
6ee13bf -
Revert "IP based rate limit"
This reverts commit 87c693736f128927291f8b396dba85fa992539ae.2e8d14d -
IP based rate limit
87c6937 -
Store IP in messageEvents
b8c0a1d
-
🐛 Fix - preprompt not being passed correctly
The default case should be to use the model preprompt which wasn't being done. Closes #4147560449
-
🐛 Fix - Remove the min length validation on preprompt
Should fix the docker container7533ab7
-
fix auth on websearch (#410)
Unverified signaturee91b76c -
🐛 Fix - Model switching no longer enables data sharing
f2c9036 -
types fix in SettingsModal.svelte
8a3914c -
fix issue with settings modal
3fd838e
-
Support custom system prompts from the user (#399)
Unverified signature* Support custom system prompts from the user * linter * types & lintcd6894d -
Make all prompt templates configurable (#400)
Unverified signature* Make all prompt templates configurable This PR enables all prompts to be configurable using handlebar templates as described in #382. For backward compatibility the old hardcoded templates are reimplemented as the default templates. The old template parameters such as `preprompt`, `userMessageToken`, `userMessageEndToken`, `assistantMessageToken`, `assistantMessageEndToken` are now considered legacy. They still work as they are exposed as variables to the default template. However, new prompt configurations should not use these. And it is recommended that the legacy support is eventually removed. As an example, this is how the default chat prompt template is implemented: ``` {{preprompt}} {{#each messages}} {{#ifUser}}{{@root.userMessageToken}}{{content}}{{@root.userMessageEndToken}}{{/ifUser}} {{#ifAssistant}}{{@root.assistantMessageToken}}{{content}}{{@root.assistantMessageEndToken}}{{/ifAssistant}} {{/each}} {{assistantMessageToken}} ``` In addition, …447c0ca
-
Add Sagemaker support (#401)
Unverified signature* work on sagemaker support * fix sagemaker integration * remove unnecessary deps * fix default endpoint * remove unneeded deps, fixed types * Use conditional validation for endpoints This was needed because the discriminated union couldn't handle the legacy case where `host` is undefined. * add note in readme about aws sagemaker2e6d1bb
-
feat: add support for endpoints requiring client authentication using PKI (#393)
TomUnverified signature* feat: add support for endpoints requiring client authentication using PKI This enables a global mTLS context for client requests using the native fetch API. It can be used when custom endpoints require client authentication via PKI. * fix: Dockerfile syntax required for macOS Set the Dockerfile syntax explictly to enable new features in Dockerfile. The default syntax uses by Docker Desktop for macOS does not include support for `--link` in the COPY command for example.66adc5d
-
Leverage model link to modelUrl when informed (#385)
Albert IribarneUnverified signature* Use modelUrl instead of building it from model name * Preserve compatibility with optional modelUrl config Use modelUrl if informed, else use the previous behavior.0ad340e
-
allow different user and assistant end-token (#375)
Unverified signature* allow different user and assistant end-token For models like Llama2, the EndToken is not the same for a userMessage and an assistantMessage. This implements `userMessageEndToken` and `assistantMessageEndToken` which overwrites the messageEndToken behavior. This PR also allows empty strings as userMessageToken and assistantMessageToken and makes this the default. This adds additional flexibility, which is required in the case of Llama2 where the first userMessage is effectively different because of the system message. Note that because `userMessageEndToken` and `assistantMessageToken` are nearly always concatenated, it is almost redundant to have both. The exception is `generateQuery` for websearch which have several consecutive user messages. * Make model branding customizable based on env var (#345) * rm open assistant branding * Update SettingsModal.svelte * make settings work with a dynamic list of models * fixed types --------- Co-authored-…f209301 -
add a login button when users are logged out (#381)
Unverified signature8fa7bd9 -
trim and remove stop-suffixes from summary (#369)
Unverified signatureThe chat generation removes parameters.stop and <|endoftext|> from the generated text. And additionally trims trailing whitespace. This PR copies that behavior to the summarize functionality, when the summary is produced by a the chat model.d2a650e
-
Make model branding customizable based on env var (#345)
Unverified signature* rm open assistant branding * Update SettingsModal.svelte * make settings work with a dynamic list of models * fixed types --------- Co-authored-by: Nathan Sarrazin <sarrazin.nathan@gmail.com>54e8a52
-
Update /privacy and other content following Llama v2 release (#374)
Unverified signature7dd8724
-
Revert "support rate limiting based on user IP (#342)"
This reverts commit 77677573c5f5a05c1c45d481089fd49e14709b10.ac291a6 -
Reopen: Feature/disable login for n messages (#356)
Unverified signature* disable login on first message * update banner here too * modal wording tweaks * prevent NaN * fix login wall * fix flicker * lint * put modal text behind login check * fix bug with sending messages without login * fix misalignment between ui and api * fix data update on disable login --------- Co-authored-by: Nathan Sarrazin <sarrazin.nathan@gmail.com>7c4fdc9 -
support rate limiting based on user IP (#342)
Unverified signature7767757
-
Revert "Option to disable login on first N messages (#352)"
This reverts commit 6183fe78b431f3b8319fa5b271ed017d2bdf47b4.0a662b7 -
Option to disable login on first N messages (#352)
Unverified signature* disable login on first message * update banner here too * modal wording tweaks * prevent NaN --------- Co-authored-by: Victor Mustar <victor.mustar@gmail.com>6183fe7