mirror of
https://github.com/danny-avila/LibreChat.git
synced 2025-12-17 00:40:14 +01:00
🪨 feat: AWS Bedrock support (#3935)
* feat: Add BedrockIcon component to SVG library * feat: EModelEndpoint.bedrock * feat: first pass, bedrock chat. note: AgentClient is returning `agents` as conversation.endpoint * fix: declare endpoint in initialization step * chore: Update @librechat/agents dependency to version 1.4.5 * feat: backend content aggregation for agents/bedrock * feat: abort agent requests * feat: AWS Bedrock icons * WIP: agent provider schema parsing * chore: Update EditIcon props type * refactor(useGenerationsByLatest): make agents and bedrock editable * refactor: non-assistant message content, parts * fix: Bedrock response `sender` * fix: use endpointOption.model_parameters not endpointOption.modelOptions * fix: types for step handler * refactor: Update Agents.ToolCallDelta type * refactor: Remove unnecessary assignment of parentMessageId in AskController * refactor: remove unnecessary assignment of parentMessageId (agent request handler) * fix(bedrock/agents): message regeneration * refactor: dynamic form elements using react-hook-form Controllers * fix: agent icons/labels for messages * fix: agent actions * fix: use of new dynamic tags causing application crash * refactor: dynamic settings touch-ups * refactor: update Slider component to allow custom track class name * refactor: update DynamicSlider component styles * refactor: use Constants value for GLOBAL_PROJECT_NAME (enum) * feat: agent share global methods/controllers * fix: agents query * fix: `getResponseModel` * fix: share prompt a11y issue * refactor: update SharePrompt dialog theme styles * refactor: explicit typing for SharePrompt * feat: add agent roles/permissions * chore: update @librechat/agents dependency to version 1.4.7 for tool_call_ids edge case * fix(Anthropic): messages.X.content.Y.tool_use.input: Input should be a valid dictionary * fix: handle text parts with tool_call_ids and empty text * fix: role initialization * refactor: don't make instructions required * refactor: improve typing of Text part * fix: setShowStopButton for agents route * chore: remove params for now * fix: add streamBuffer and streamRate to help prevent 'Overloaded' errors from Anthropic API * refactor: remove console.log statement in ContentRender component * chore: typing, rename Context to Delete Button * chore(DeleteButton): logging * refactor(Action): make accessible * style(Action): improve a11y again * refactor: remove use/mention of mongoose sessions * feat: first pass, sharing agents * feat: visual indicator for global agent, remove author when serving to non-author * wip: params * chore: fix typing issues * fix(schemas): typing * refactor: improve accessibility of ListCard component and fix console React warning * wip: reset templates for non-legacy new convos * Revert "wip: params" This reverts commitf8067e91d4. * Revert "refactor: dynamic form elements using react-hook-form Controllers" This reverts commit2150c4815d. * fix(Parameters): types and parameter effect update to only update local state to parameters * refactor: optimize useDebouncedInput hook for better performance * feat: first pass, anthropic bedrock params * chore: paramEndpoints check for endpointType too * fix: maxTokens to use coerceNumber.optional(), * feat: extra chat model params * chore: reduce code repetition * refactor: improve preset title handling in SaveAsPresetDialog component * refactor: improve preset handling in HeaderOptions component * chore: improve typing, replace legacy dialog for SaveAsPresetDialog * feat: save as preset from parameters panel * fix: multi-search in select dropdown when using Option type * refactor: update default showDefault value to false in Dynamic components * feat: Bedrock presets settings * chore: config, fix agents schema, update config version * refactor: update AWS region variable name in bedrock options endpoint to BEDROCK_AWS_DEFAULT_REGION * refactor: update baseEndpointSchema in config.ts to include baseURL property * refactor: update createRun function to include req parameter and set streamRate based on provider * feat: availableRegions via config * refactor: remove unused demo agent controller file * WIP: title * Update @librechat/agents to version 1.5.0 * chore: addTitle.js to handle empty responseText * feat: support images and titles * feat: context token updates * Refactor BaseClient test to use expect.objectContaining * refactor: add model select, remove header options params, move side panel params below prompts * chore: update models list, catch title error * feat: model service for bedrock models (env) * chore: Remove verbose debug log in AgentClient class following stream * feat(bedrock): track token spend; fix: token rates, value key mapping for AWS models * refactor: handle streamRate in `handleLLMNewToken` callback * chore: AWS Bedrock example config in `.env.example` * refactor: Rename bedrockMeta to bedrockGeneral in settings.ts and use for AI21 and Amazon Bedrock providers * refactor: Update `.env.example` with AWS Bedrock model IDs URL and additional notes * feat: titleModel support for bedrock * refactor: Update `.env.example` with additional notes for AWS Bedrock model IDs
This commit is contained in:
Danny Avila
GitHub
parent
8c14360263
commit
d59b62174f
134 changed files with 3684 additions and 1213 deletions
|
|
@ -7,9 +7,11 @@
|
|||
// validateVisionModel,
|
||||
// mapModelToAzureConfig,
|
||||
// } = require('librechat-data-provider');
|
||||
const { Callback } = require('@librechat/agents');
|
||||
const { Callback, createMetadataAggregator } = require('@librechat/agents');
|
||||
const {
|
||||
Constants,
|
||||
EModelEndpoint,
|
||||
bedrockOutputParser,
|
||||
providerEndpointMap,
|
||||
removeNullishValues,
|
||||
} = require('librechat-data-provider');
|
||||
|
|
@ -23,15 +25,27 @@ const {
|
|||
formatAgentMessages,
|
||||
createContextHandlers,
|
||||
} = require('~/app/clients/prompts');
|
||||
const { encodeAndFormat } = require('~/server/services/Files/images/encode');
|
||||
const Tokenizer = require('~/server/services/Tokenizer');
|
||||
const { spendTokens } = require('~/models/spendTokens');
|
||||
const BaseClient = require('~/app/clients/BaseClient');
|
||||
// const { sleep } = require('~/server/utils');
|
||||
const { createRun } = require('./run');
|
||||
const { logger } = require('~/config');
|
||||
|
||||
/** @typedef {import('@librechat/agents').MessageContentComplex} MessageContentComplex */
|
||||
|
||||
// const providerSchemas = {
|
||||
// [EModelEndpoint.bedrock]: true,
|
||||
// };
|
||||
|
||||
const providerParsers = {
|
||||
[EModelEndpoint.bedrock]: bedrockOutputParser,
|
||||
};
|
||||
|
||||
class AgentClient extends BaseClient {
|
||||
constructor(options = {}) {
|
||||
super(options);
|
||||
super(null, options);
|
||||
|
||||
/** @type {'discard' | 'summarize'} */
|
||||
this.contextStrategy = 'discard';
|
||||
|
|
@ -39,11 +53,31 @@ class AgentClient extends BaseClient {
|
|||
/** @deprecated @type {true} - Is a Chat Completion Request */
|
||||
this.isChatCompletion = true;
|
||||
|
||||
const { maxContextTokens, modelOptions = {}, ...clientOptions } = options;
|
||||
/** @type {AgentRun} */
|
||||
this.run;
|
||||
|
||||
const {
|
||||
maxContextTokens,
|
||||
modelOptions = {},
|
||||
contentParts,
|
||||
collectedUsage,
|
||||
...clientOptions
|
||||
} = options;
|
||||
|
||||
this.modelOptions = modelOptions;
|
||||
this.maxContextTokens = maxContextTokens;
|
||||
this.options = Object.assign({ endpoint: EModelEndpoint.agents }, clientOptions);
|
||||
/** @type {MessageContentComplex[]} */
|
||||
this.contentParts = contentParts;
|
||||
/** @type {Array<UsageMetadata>} */
|
||||
this.collectedUsage = collectedUsage;
|
||||
this.options = Object.assign({ endpoint: options.endpoint }, clientOptions);
|
||||
}
|
||||
|
||||
/**
|
||||
* Returns the aggregated content parts for the current run.
|
||||
* @returns {MessageContentComplex[]} */
|
||||
getContentParts() {
|
||||
return this.contentParts;
|
||||
}
|
||||
|
||||
setOptions(options) {
|
||||
|
|
@ -112,9 +146,27 @@ class AgentClient extends BaseClient {
|
|||
}
|
||||
|
||||
getSaveOptions() {
|
||||
const parseOptions = providerParsers[this.options.endpoint];
|
||||
let runOptions =
|
||||
this.options.endpoint === EModelEndpoint.agents
|
||||
? {
|
||||
model: undefined,
|
||||
// TODO:
|
||||
// would need to be override settings; otherwise, model needs to be undefined
|
||||
// model: this.override.model,
|
||||
// instructions: this.override.instructions,
|
||||
// additional_instructions: this.override.additional_instructions,
|
||||
}
|
||||
: {};
|
||||
|
||||
if (parseOptions) {
|
||||
runOptions = parseOptions(this.modelOptions);
|
||||
}
|
||||
|
||||
return removeNullishValues(
|
||||
Object.assign(
|
||||
{
|
||||
endpoint: this.options.endpoint,
|
||||
agent_id: this.options.agent.id,
|
||||
modelLabel: this.options.modelLabel,
|
||||
maxContextTokens: this.options.maxContextTokens,
|
||||
|
|
@ -122,15 +174,8 @@ class AgentClient extends BaseClient {
|
|||
imageDetail: this.options.imageDetail,
|
||||
spec: this.options.spec,
|
||||
},
|
||||
this.modelOptions,
|
||||
{
|
||||
model: undefined,
|
||||
// TODO:
|
||||
// would need to be override settings; otherwise, model needs to be undefined
|
||||
// model: this.override.model,
|
||||
// instructions: this.override.instructions,
|
||||
// additional_instructions: this.override.additional_instructions,
|
||||
},
|
||||
// TODO: PARSE OPTIONS BY PROVIDER, MAY CONTAIN SENSITIVE DATA
|
||||
runOptions,
|
||||
),
|
||||
);
|
||||
}
|
||||
|
|
@ -142,6 +187,16 @@ class AgentClient extends BaseClient {
|
|||
};
|
||||
}
|
||||
|
||||
async addImageURLs(message, attachments) {
|
||||
const { files, image_urls } = await encodeAndFormat(
|
||||
this.options.req,
|
||||
attachments,
|
||||
this.options.agent.provider,
|
||||
);
|
||||
message.image_urls = image_urls.length ? image_urls : undefined;
|
||||
return files;
|
||||
}
|
||||
|
||||
async buildMessages(
|
||||
messages,
|
||||
parentMessageId,
|
||||
|
|
@ -270,25 +325,34 @@ class AgentClient extends BaseClient {
|
|||
/** @type {sendCompletion} */
|
||||
async sendCompletion(payload, opts = {}) {
|
||||
this.modelOptions.user = this.user;
|
||||
return await this.chatCompletion({
|
||||
await this.chatCompletion({
|
||||
payload,
|
||||
onProgress: opts.onProgress,
|
||||
abortController: opts.abortController,
|
||||
});
|
||||
return this.contentParts;
|
||||
}
|
||||
|
||||
// async recordTokenUsage({ promptTokens, completionTokens, context = 'message' }) {
|
||||
// await spendTokens(
|
||||
// {
|
||||
// context,
|
||||
// model: this.modelOptions.model,
|
||||
// conversationId: this.conversationId,
|
||||
// user: this.user ?? this.options.req.user?.id,
|
||||
// endpointTokenConfig: this.options.endpointTokenConfig,
|
||||
// },
|
||||
// { promptTokens, completionTokens },
|
||||
// );
|
||||
// }
|
||||
/**
|
||||
* @param {Object} params
|
||||
* @param {string} [params.model]
|
||||
* @param {string} [params.context='message']
|
||||
* @param {UsageMetadata[]} [params.collectedUsage=this.collectedUsage]
|
||||
*/
|
||||
async recordCollectedUsage({ model, context = 'message', collectedUsage = this.collectedUsage }) {
|
||||
for (const usage of collectedUsage) {
|
||||
await spendTokens(
|
||||
{
|
||||
context,
|
||||
model: model ?? this.modelOptions.model,
|
||||
conversationId: this.conversationId,
|
||||
user: this.user ?? this.options.req.user?.id,
|
||||
endpointTokenConfig: this.options.endpointTokenConfig,
|
||||
},
|
||||
{ promptTokens: usage.input_tokens, completionTokens: usage.output_tokens },
|
||||
);
|
||||
}
|
||||
}
|
||||
|
||||
async chatCompletion({ payload, abortController = null }) {
|
||||
try {
|
||||
|
|
@ -398,9 +462,8 @@ class AgentClient extends BaseClient {
|
|||
// });
|
||||
// }
|
||||
|
||||
// const streamRate = this.options.streamRate ?? Constants.DEFAULT_STREAM_RATE;
|
||||
|
||||
const run = await createRun({
|
||||
req: this.options.req,
|
||||
agent: this.options.agent,
|
||||
tools: this.options.tools,
|
||||
toolMap: this.options.toolMap,
|
||||
|
|
@ -415,6 +478,7 @@ class AgentClient extends BaseClient {
|
|||
thread_id: this.conversationId,
|
||||
},
|
||||
run_id: this.responseMessageId,
|
||||
signal: abortController.signal,
|
||||
streamMode: 'values',
|
||||
version: 'v2',
|
||||
};
|
||||
|
|
@ -423,8 +487,10 @@ class AgentClient extends BaseClient {
|
|||
throw new Error('Failed to create run');
|
||||
}
|
||||
|
||||
this.run = run;
|
||||
|
||||
const messages = formatAgentMessages(payload);
|
||||
const runMessages = await run.processStream({ messages }, config, {
|
||||
await run.processStream({ messages }, config, {
|
||||
[Callback.TOOL_ERROR]: (graph, error, toolId) => {
|
||||
logger.error(
|
||||
'[api/server/controllers/agents/client.js #chatCompletion] Tool Error',
|
||||
|
|
@ -433,14 +499,94 @@ class AgentClient extends BaseClient {
|
|||
);
|
||||
},
|
||||
});
|
||||
// console.dir(runMessages, { depth: null });
|
||||
return runMessages;
|
||||
this.recordCollectedUsage({ context: 'message' }).catch((err) => {
|
||||
logger.error(
|
||||
'[api/server/controllers/agents/client.js #chatCompletion] Error recording collected usage',
|
||||
err,
|
||||
);
|
||||
});
|
||||
} catch (err) {
|
||||
logger.error(
|
||||
'[api/server/controllers/agents/client.js #chatCompletion] Unhandled error type',
|
||||
if (!abortController.signal.aborted) {
|
||||
logger.error(
|
||||
'[api/server/controllers/agents/client.js #sendCompletion] Unhandled error type',
|
||||
err,
|
||||
);
|
||||
throw err;
|
||||
}
|
||||
|
||||
logger.warn(
|
||||
'[api/server/controllers/agents/client.js #sendCompletion] Operation aborted',
|
||||
err,
|
||||
);
|
||||
throw err;
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
*
|
||||
* @param {Object} params
|
||||
* @param {string} params.text
|
||||
* @param {string} params.conversationId
|
||||
*/
|
||||
async titleConvo({ text }) {
|
||||
if (!this.run) {
|
||||
throw new Error('Run not initialized');
|
||||
}
|
||||
const { handleLLMEnd, collected: collectedMetadata } = createMetadataAggregator();
|
||||
const clientOptions = {};
|
||||
const providerConfig = this.options.req.app.locals[this.options.agent.provider];
|
||||
if (
|
||||
providerConfig &&
|
||||
providerConfig.titleModel &&
|
||||
providerConfig.titleModel !== Constants.CURRENT_MODEL
|
||||
) {
|
||||
clientOptions.model = providerConfig.titleModel;
|
||||
}
|
||||
try {
|
||||
const titleResult = await this.run.generateTitle({
|
||||
inputText: text,
|
||||
contentParts: this.contentParts,
|
||||
clientOptions,
|
||||
chainOptions: {
|
||||
callbacks: [
|
||||
{
|
||||
handleLLMEnd,
|
||||
},
|
||||
],
|
||||
},
|
||||
});
|
||||
|
||||
const collectedUsage = collectedMetadata.map((item) => {
|
||||
let input_tokens, output_tokens;
|
||||
|
||||
if (item.usage) {
|
||||
input_tokens = item.usage.input_tokens || item.usage.inputTokens;
|
||||
output_tokens = item.usage.output_tokens || item.usage.outputTokens;
|
||||
} else if (item.tokenUsage) {
|
||||
input_tokens = item.tokenUsage.promptTokens;
|
||||
output_tokens = item.tokenUsage.completionTokens;
|
||||
}
|
||||
|
||||
return {
|
||||
input_tokens: input_tokens,
|
||||
output_tokens: output_tokens,
|
||||
};
|
||||
});
|
||||
|
||||
this.recordCollectedUsage({
|
||||
model: clientOptions.model,
|
||||
context: 'title',
|
||||
collectedUsage,
|
||||
}).catch((err) => {
|
||||
logger.error(
|
||||
'[api/server/controllers/agents/client.js #titleConvo] Error recording collected usage',
|
||||
err,
|
||||
);
|
||||
});
|
||||
|
||||
return titleResult.title;
|
||||
} catch (err) {
|
||||
logger.error('[api/server/controllers/agents/client.js #titleConvo] Error', err);
|
||||
return;
|
||||
}
|
||||
}
|
||||
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue