mirror of
https://github.com/danny-avila/LibreChat.git
synced 2025-09-22 06:00:56 +02:00
4 commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Danny Avila
|
365c39c405
|
feat: Accurate Token Usage Tracking & Optional Balance (#1018)
* refactor(Chains/llms): allow passing callbacks * refactor(BaseClient): accurately count completion tokens as generation only * refactor(OpenAIClient): remove unused getTokenCountForResponse, pass streaming var and callbacks in initializeLLM * wip: summary prompt tokens * refactor(summarizeMessages): new cut-off strategy that generates a better summary by adding context from beginning, truncating the middle, and providing the end wip: draft out relevant providers and variables for token tracing * refactor(createLLM): make streaming prop false by default * chore: remove use of getTokenCountForResponse * refactor(agents): use BufferMemory as ConversationSummaryBufferMemory token usage not easy to trace * chore: remove passing of streaming prop, also console log useful vars for tracing * feat: formatFromLangChain helper function to count tokens for ChatModelStart * refactor(initializeLLM): add role for LLM tracing * chore(formatFromLangChain): update JSDoc * feat(formatMessages): formats langChain messages into OpenAI payload format * chore: install openai-chat-tokens * refactor(formatMessage): optimize conditional langChain logic fix(formatFromLangChain): fix destructuring * feat: accurate prompt tokens for ChatModelStart before generation * refactor(handleChatModelStart): move to callbacks dir, use factory function * refactor(initializeLLM): rename 'role' to 'context' * feat(Balance/Transaction): new schema/models for tracking token spend refactor(Key): factor out model export to separate file * refactor(initializeClient): add req,res objects to client options * feat: add-balance script to add to an existing users' token balance refactor(Transaction): use multiplier map/function, return balance update * refactor(Tx): update enum for tokenType, return 1 for multiplier if no map match * refactor(Tx): add fair fallback value multiplier incase the config result is undefined * refactor(Balance): rename 'tokens' to 'tokenCredits' * feat: balance check, add tx.js for new tx-related methods and tests * chore(summaryPrompts): update prompt token count * refactor(callbacks): pass req, res wip: check balance * refactor(Tx): make convoId a String type, fix(calculateTokenValue) * refactor(BaseClient): add conversationId as client prop when assigned * feat(RunManager): track LLM runs with manager, track token spend from LLM, refactor(OpenAIClient): use RunManager to create callbacks, pass user prop to langchain api calls * feat(spendTokens): helper to spend prompt/completion tokens * feat(checkBalance): add helper to check, log, deny request if balance doesn't have enough funds refactor(Balance): static check method to return object instead of boolean now wip(OpenAIClient): implement use of checkBalance * refactor(initializeLLM): add token buffer to assure summary isn't generated when subsequent payload is too large refactor(OpenAIClient): add checkBalance refactor(createStartHandler): add checkBalance * chore: remove prompt and completion token logging from route handler * chore(spendTokens): add JSDoc * feat(logTokenCost): record transactions for basic api calls * chore(ask/edit): invoke getResponseSender only once per API call * refactor(ask/edit): pass promptTokens to getIds and include in abort data * refactor(getIds -> getReqData): rename function * refactor(Tx): increase value if incomplete message * feat: record tokenUsage when message is aborted * refactor: subtract tokens when payload includes function_call * refactor: add namespace for token_balance * fix(spendTokens): only execute if corresponding token type amounts are defined * refactor(checkBalance): throws Error if not enough token credits * refactor(runTitleChain): pass and use signal, spread object props in create helpers, and use 'call' instead of 'run' * fix(abortMiddleware): circular dependency, and default to empty string for completionTokens * fix: properly cancel title requests when there isn't enough tokens to generate * feat(predictNewSummary): custom chain for summaries to allow signal passing refactor(summaryBuffer): use new custom chain * feat(RunManager): add getRunByConversationId method, refactor: remove run and throw llm error on handleLLMError * refactor(createStartHandler): if summary, add error details to runs * fix(OpenAIClient): support aborting from summarization & showing error to user refactor(summarizeMessages): remove unnecessary operations counting summaryPromptTokens and note for alternative, pass signal to summaryBuffer * refactor(logTokenCost -> recordTokenUsage): rename * refactor(checkBalance): include promptTokens in errorMessage * refactor(checkBalance/spendTokens): move to models dir * fix(createLanguageChain): correctly pass config * refactor(initializeLLM/title): add tokenBuffer of 150 for balance check * refactor(openAPIPlugin): pass signal and memory, filter functions by the one being called * refactor(createStartHandler): add error to run if context is plugins as well * refactor(RunManager/handleLLMError): throw error immediately if plugins, don't remove run * refactor(PluginsClient): pass memory and signal to tools, cleanup error handling logic * chore: use absolute equality for addTitle condition * refactor(checkBalance): move checkBalance to execute after userMessage and tokenCounts are saved, also make conditional * style: icon changes to match official * fix(BaseClient): getTokenCountForResponse -> getTokenCount * fix(formatLangChainMessages): add kwargs as fallback prop from lc_kwargs, update JSDoc * refactor(Tx.create): does not update balance if CHECK_BALANCE is not enabled * fix(e2e/cleanUp): cleanup new collections, import all model methods from index * fix(config/add-balance): add uncaughtException listener * fix: circular dependency * refactor(initializeLLM/checkBalance): append new generations to errorMessage if cost exceeds balance * fix(handleResponseMessage): only record token usage in this method if not error and completion is not skipped * fix(createStartHandler): correct condition for generations * chore: bump postcss due to moderate severity vulnerability * chore: bump zod due to low severity vulnerability * chore: bump openai & data-provider version * feat(types): OpenAI Message types * chore: update bun lockfile * refactor(CodeBlock): add error block formatting * refactor(utils/Plugin): factor out formatJSON and cn to separate files (json.ts and cn.ts), add extractJSON * chore(logViolation): delete user_id after error is logged * refactor(getMessageError -> Error): change to React.FC, add token_balance handling, use extractJSON to determine JSON instead of regex * fix(DALL-E): use latest openai SDK * chore: reorganize imports, fix type issue * feat(server): add balance route * fix(api/models): add auth * feat(data-provider): /api/balance query * feat: show balance if checking is enabled, refetch on final message or error * chore: update docs, .env.example with token_usage info, add balance script command * fix(Balance): fallback to empty obj for balance query * style: slight adjustment of balance element * docs(token_usage): add PR notes |
||
|
Danny Avila
|
317a1bd8da
|
feat: ConversationSummaryBufferMemory (#973)
* refactor: pass model in message edit payload, use encoder in standalone util function * feat: add summaryBuffer helper * refactor(api/messages): use new countTokens helper and add auth middleware at top * wip: ConversationSummaryBufferMemory * refactor: move pre-generation helpers to prompts dir * chore: remove console log * chore: remove test as payload will no longer carry tokenCount * chore: update getMessagesWithinTokenLimit JSDoc * refactor: optimize getMessagesForConversation and also break on summary, feat(ci): getMessagesForConversation tests * refactor(getMessagesForConvo): count '00000000-0000-0000-0000-000000000000' as root message * chore: add newer model to token map * fix: condition was point to prop of array instead of message prop * refactor(BaseClient): use object for refineMessages param, rename 'summary' to 'summaryMessage', add previous_summary refactor(getMessagesWithinTokenLimit): replace text and tokenCount if should summarize, summary, and summaryTokenCount are present fix/refactor(handleContextStrategy): use the right comparison length for context diff, and replace payload first message when a summary is present * chore: log previous_summary if debugging * refactor(formatMessage): assume if role is defined that it's a valid value * refactor(getMessagesWithinTokenLimit): remove summary logic refactor(handleContextStrategy): add usePrevSummary logic in case only summary was pruned refactor(loadHistory): initial message query will return all ordered messages but keep track of the latest summary refactor(getMessagesForConversation): use object for single param, edit jsdoc, edit all files using the method refactor(ChatGPTClient): order messages before buildPrompt is called, TODO: add convoSumBuffMemory logic * fix: undefined handling and summarizing only when shouldRefineContext is true * chore(BaseClient): fix test results omitting system role for summaries and test edge case * chore: export summaryBuffer from index file * refactor(OpenAIClient/BaseClient): move refineMessages to subclass, implement LLM initialization for summaryBuffer * feat: add OPENAI_SUMMARIZE to enable summarizing, refactor: rename client prop 'shouldRefineContext' to 'shouldSummarize', change contextStrategy value to 'summarize' from 'refine' * refactor: rename refineMessages method to summarizeMessages for clarity * chore: clarify summary future intent in .env.example * refactor(initializeLLM): handle case for either 'model' or 'modelName' being passed * feat(gptPlugins): enable summarization for plugins * refactor(gptPlugins): utilize new initializeLLM method and formatting methods for messages, use payload array for currentMessages and assign pastMessages sooner * refactor(agents): use ConversationSummaryBufferMemory for both agent types * refactor(formatMessage): optimize original method for langchain, add helper function for langchain messages, add JSDocs and tests * refactor(summaryBuffer): add helper to createSummaryBufferMemory, and use new formatting helpers * fix: forgot to spread formatMessages also took opportunity to pluralize filename * refactor: pass memory to tools, namely openapi specs. not used and may never be used by new method but added for testing * ci(formatMessages): add more exhaustive checks for langchain messages * feat: add debug env var for OpenAI * chore: delete unnecessary comments * chore: add extra note about summary feature * fix: remove tokenCount from payload instructions * fix: test fail * fix: only pass instructions to payload when defined or not empty object * refactor: fromPromptMessages is deprecated, use renamed method fromMessages * refactor: use 'includes' instead of 'startsWith' for extended OpenRouter compatibility * fix(PluginsClient.buildPromptBody): handle undefined message strings * chore: log langchain titling error * feat: getModelMaxTokens helper * feat: tokenSplit helper * feat: summary prompts updated * fix: optimize _CUT_OFF_SUMMARIZER prompt * refactor(summaryBuffer): use custom summary prompt, allow prompt to be passed, pass humanPrefix and aiPrefix to memory, along with any future variables, rename messagesToRefine to context * fix(summaryBuffer): handle edge case where messagesToRefine exceeds summary context, refactor(BaseClient): allow custom maxContextTokens to be passed to getMessagesWithinTokenLimit, add defined check before unshifting summaryMessage, update shouldSummarize based on this refactor(OpenAIClient): use getModelMaxTokens, use cut-off message method for summary if no messages were left after pruning * fix(handleContextStrategy): handle case where incoming prompt is bigger than model context * chore: rename refinedContent to splitText * chore: remove unnecessary debug log |
||
|
Danny Avila
|
e5336039fc
|
ci(backend-review.yml): add linter step to the backend review workflow (#625)
* ci(backend-review.yml): add linter step to the backend review workflow * chore(backend-review.yml): remove prettier from lint-action configuration * chore: apply new linting workflow * chore(lint-staged.config.js): reorder lint-staged tasks for JavaScript and TypeScript files * chore(eslint): update ignorePatterns in .eslintrc.js chore(lint-action): remove prettier option in backend-review.yml chore(package.json): add lint and lint:fix scripts * chore(lint-staged.config.js): remove prettier --write command for js, jsx, ts, tsx files * chore(titleConvo.js): remove unnecessary console.log statement chore(titleConvo.js): add missing comma in options object * chore: apply linting to all files * chore(lint-staged.config.js): update lint-staged configuration to include prettier formatting |
||
|
Danny Avila
|
8819e83d2c
|
refactor: Client Classes & Azure OpenAI as a separate Endpoint (#532)
* refactor: start new client classes, test localAi support * feat: create base class, extend chatgpt from base * refactor(BaseClient.js): change userId parameter to user refactor(BaseClient.js): change userId parameter to user feat(OpenAIClient.js): add sendMessage method refactor(OpenAIClient.js): change getConversation method to use user parameter instead of userId refactor(OpenAIClient.js): change saveMessageToDatabase method to use user parameter instead of userId refactor(OpenAIClient.js): change buildPrompt method to use messages parameter instead of orderedMessages feat(index.js): export client classes refactor(askGPTPlugins.js): use req.body.token or process.env.OPENAI_API_KEY as OpenAI API key refactor(index.js): comment out askOpenAI route feat(index.js): add openAI route feat(openAI.js): add new route for OpenAI API requests with support for progress updates and aborting requests. * refactor(BaseClient.js): use optional chaining operator to access messageId property refactor(OpenAIClient.js): use orderedMessages instead of messages to build prompt refactor(OpenAIClient.js): use optional chaining operator to access messageId property refactor(fetch-polyfill.js): remove fetch polyfill refactor(openAI.js): comment out debug option in clientOptions * refactor: update import statements and remove unused imports in several files feat: add getAzureCredentials function to azureUtils module docs: update comments in azureUtils module * refactor(utils): rename migrateConversations to migrateDataToFirstUser for clarity and consistency * feat(chatgpt-client.js): add getAzureCredentials function to retrieve Azure credentials feat(chatgpt-client.js): use getAzureCredentials function to generate reverseProxyUrl feat(OpenAIClient.js): add isChatCompletion property to determine if chat completion model is used feat(OpenAIClient.js): add saveOptions parameter to sendMessage and buildPrompt methods feat(OpenAIClient.js): modify buildPrompt method to handle chat completion model feat(openAI.js): modify endpointOption to include modelOptions instead of individual options refactor(OpenAIClient.js): modify getDelta property to use isChatCompletion property instead of isChatGptModel property refactor(OpenAIClient.js): modify sendMessage method to use saveOptions parameter instead of modelOptions parameter refactor(OpenAIClient.js): modify buildPrompt method to use saveOptions parameter instead of modelOptions parameter refactor(OpenAIClient.js): modify ask method to include endpointOption parameter * chore: delete draft file * refactor(OpenAIClient.js): extract sendCompletion method from sendMessage method for reusability * refactor(BaseClient.js): move sendMessage method to BaseClient class feat(OpenAIClient.js): inherit from BaseClient class and implement necessary methods and properties for OpenAIClient class. * refactor(BaseClient.js): rename getBuildPromptOptions to getBuildMessagesOptions feat(BaseClient.js): add buildMessages method to BaseClient class fix(ChatGPTClient.js): use message.text instead of message.message refactor(ChatGPTClient.js): rename buildPromptBody to buildMessagesBody refactor(ChatGPTClient.js): remove console.debug statement and add debug log for prompt variable refactor(OpenAIClient.js): move setOptions method to the bottom of the class feat(OpenAIClient.js): add support for cl100k_base encoding feat(OpenAIClient.js): add support for unofficial chat GPT models feat(OpenAIClient.js): add support for custom modelOptions feat(OpenAIClient.js): add caching for tokenizers feat(OpenAIClient.js): add freeAndInitializeEncoder method to free and reinitialize tokenizers refactor(OpenAIClient.js): rename getBuildPromptOptions to getBuildMessagesOptions refactor(OpenAIClient.js): rename buildPrompt to buildMessages refactor(OpenAIClient.js): remove endpointOption from ask function arguments in openAI.js * refactor(ChatGPTClient.js, OpenAIClient.js): improve code readability and consistency - In ChatGPTClient.js, update the roleLabel and messageString variables to handle cases where the message object does not have an isCreatedByUser property or a role property with a value of 'user'. - In OpenAIClient.js, rename the freeAndInitializeEncoder method to freeAndResetEncoder to better reflect its functionality. Also, update the method calls to reflect the new name. Additionally, update the getTokenCount method to handle errors by calling the freeAndResetEncoder method instead of the now-renamed freeAndInitializeEncoder method. * refactor(OpenAIClient.js): extract instructions object to a separate variable and add it to payload after formatted messages fix(OpenAIClient.js): handle cases where progressMessage.choices is undefined or empty * refactor(BaseClient.js): extract addInstructions method from sendMessage method feat(OpenAIClient.js): add maxTokensMap object to map maximum tokens for each model refactor(OpenAIClient.js): use addInstructions method in buildMessages method instead of manually building the payload list * refactor(OpenAIClient.js): remove unnecessary condition for modelOptions.model property in buildMessages method * feat(BaseClient.js): add support for token count tracking and context strategy feat(OpenAIClient.js): add support for token count tracking and context strategy feat(Message.js): add tokenCount field to Message schema and updateMessage function * refactor(BaseClient.js): add support for refining messages based on token limit feat(OpenAIClient.js): add support for context refinement strategy refactor(OpenAIClient.js): use context refinement strategy in message sending refactor(server/index.js): improve code readability by breaking long lines * refactor(BaseClient.js): change `remainingContext` to `remainingContextTokens` for clarity feat(BaseClient.js): add `refinePrompt` and `refinePromptTemplate` to handle message refinement feat(BaseClient.js): add `refineMessages` method to refine messages feat(BaseClient.js): add `handleContextStrategy` method to handle context strategy feat(OpenAIClient.js): add `abortController` to `buildPrompt` method options refactor(OpenAIClient.js): change `payload` and `tokenCountMap` to let variables in `handleContextStrategy` method refactor(BaseClient.js): change `remainingContext` to `remainingContextTokens` in `handleContextStrategy` method for consistency refactor(BaseClient.js): change `remainingContext` to `remainingContextTokens` in `getMessagesWithinTokenLimit` method for consistency refactor(BaseClient.js): change `remainingContext` to `remainingContext * chore(openAI.js): comment out contextStrategy option in clientOptions * chore(openAI.js): comment out debug option in clientOptions object * test: BaseClient tests in progress * test: Complete OpenAIClient & BaseClient tests * fix(OpenAIClient.js): remove unnecessary whitespace fix(OpenAIClient.js): remove unused variables and comments fix(OpenAIClient.test.js): combine getTokenCount and freeAndResetEncoder tests * chore(.eslintrc.js): add rule for maximum of 1 empty line feat(ask/openAI.js): add abortMessage utility function fix(ask/openAI.js): handle error and abort message if partial text is less than 2 characters feat(utils/index.js): export abortMessage utility function * test: complete additional tests * feat: Azure OpenAI as a separate endpoint * chore: remove extraneous console logs * fix(azureOpenAI): use chatCompletion endpoint * chore(initializeClient.js): delete initializeClient.js file chore(askOpenAI.js): delete old OpenAI route handler chore(handlers.js): remove trailing whitespace in thought variable assignment * chore(chatgpt-client.js): remove unused chatgpt-client.js file refactor(index.js): remove askClient import and export from index.js * chore(chatgpt-client.tokens.js): update test script for memory usage and encoding performance The test script in `chatgpt-client.tokens.js` has been updated to measure the memory usage and encoding performance of the client. The script now includes information about the initial memory usage, peak memory usage, final memory usage, and memory usage after a timeout. It also provides insights into the number of encoding requests that can be processed per second. The script has been modified to use the `OpenAIClient` class instead of the `ChatGPTClient` class. Additionally, the number of iterations for the encoding loop has been reduced to 10,000. A timeout function has been added to simulate a delay of 15 seconds. After the timeout, the memory usage is measured again. The script now handles uncaught exceptions and logs any errors that occur, except for errors related to failed fetch requests. Note: This is a test script and should not be used in production * feat(FakeClient.js): add a new class `FakeClient` that extends `BaseClient` and implements methods for a fake client feat(FakeClient.js): implement the `setOptions` method to handle options for the fake client feat(FakeClient.js): implement the `initializeFakeClient` function to initialize a fake client with options and fake messages fix(OpenAIClient.js): remove duplicate `maxTokensMap` import and use the one from utils feat(BaseClient): return promptTokens and completionTokens * refactor(gptPlugins): refactor ChatAgent to PluginsClient, which extends OpenAIClient * refactor: client paths * chore(jest.config.js): remove jest.config.js file * fix(PluginController.js): update file path to manifest.json feat(gptPlugins.js): add support for aborting messages refactor(ask/index.js): rename askGPTPlugins to gptPlugins for consistency * fix(BaseClient.js): fix spacing in generateTextStream function signature refactor(BaseClient.js): remove unnecessary push to currentMessages in generateUserMessage function refactor(BaseClient.js): remove unnecessary push to currentMessages in handleStartMethods function refactor(PluginsClient.js): remove unused variables and date formatting in constructor refactor(PluginsClient.js): simplify mapping of pastMessages in getCompletionPayload function * refactor(GoogleClient): GoogleClient now extends BaseClient * chore(.env.example): add AZURE_OPENAI_MODELS variable fix(api/routes/ask/gptPlugins.js): enable Azure integration if PLUGINS_USE_AZURE is true fix(api/routes/endpoints.js): getOpenAIModels function now accepts options, use AZURE_OPENAI_MODELS if PLUGINS_USE_AZURE is true fix(client/components/Endpoints/OpenAI/Settings.jsx): remove console.log statement docs(features/azure.md): add documentation for Azure OpenAI integration and environment variables * fix(e2e:popup): includes the icon + endpoint names in role, name property |
Renamed from api/app/langchain/agents/CustomAgent/initializeCustomAgent.js (Browse further)