mirror of
https://github.com/danny-avila/LibreChat.git
synced 2025-12-17 00:40:14 +01:00
🎚️ feat: Anthropic Parameter Set Support via Custom Endpoints (#9415)
* refactor: modularize openai llm config logic into new getOpenAILLMConfig function (#9412) * ✈️ refactor: Migrate Anthropic's getLLMConfig to TypeScript (#9413) * refactor: move tokens.js over to packages/api and update imports * refactor: port tokens.js to typescript * refactor: move helpers.js over to packages/api and update imports * refactor: port helpers.js to typescript * refactor: move anthropic/llm.js over to packages/api and update imports * refactor: port anthropic/llm.js to typescript with supporting types in types/anthropic.ts and updated tests in llm.spec.js * refactor: move llm.spec.js over to packages/api and update import * refactor: port llm.spec.js over to typescript * 📝 Add Prompt Parameter Support for Anthropic Custom Endpoints (#9414) feat: add anthropic llm config support for openai-like (custom) endpoints * fix: missed compiler / type issues from addition of getAnthropicLLMConfig * refactor: update tokens.ts to export constants and functions, enhance type definitions, and adjust default values * WIP: first pass, decouple `llmConfig` from `configOptions` * chore: update import path for OpenAI configuration from 'llm' to 'config' * refactor: enhance type definitions for ThinkingConfig and update modelOptions in AnthropicConfigOptions * refactor: cleanup type, introduce openai transform from alt provider * chore: integrate removeNullishValues in Google llmConfig and update OpenAI exports * chore: bump version of @librechat/api to 1.3.5 in package.json and package-lock.json * refactor: update customParams type in OpenAIConfigOptions to use TConfig['customParams'] * refactor: enhance transformToOpenAIConfig to include fromEndpoint and improve config extraction * refactor: conform userId field for anthropic/openai, cleanup anthropic typing * ci: add backward compatibility tests for getOpenAIConfig with various endpoints and configurations * ci: replace userId with user in clientOptions for getLLMConfig * test: add Azure OpenAI endpoint tests for various configurations in getOpenAIConfig * refactor: defaultHeaders retrieval for prompt caching for anthropic-based custom endpoint (litellm) * test: add unit tests for getOpenAIConfig with various Anthropic model configurations * test: enhance Anthropic compatibility tests with addParams and dropParams handling * chore: update @librechat/agents dependency to version 2.4.78 in package.json and package-lock.json * chore: update @librechat/agents dependency to version 2.4.79 in package.json and package-lock.json --------- Co-authored-by: Danny Avila <danny@librechat.ai>
This commit is contained in:
Dustin Healy
GitHub
parent
7de6f6e44c
commit
c6ecf0095b
40 changed files with 1736 additions and 432 deletions
|
|
@ -10,7 +10,17 @@ const {
|
|||
validateVisionModel,
|
||||
} = require('librechat-data-provider');
|
||||
const { SplitStreamHandler: _Handler } = require('@librechat/agents');
|
||||
const { Tokenizer, createFetch, createStreamEventHandlers } = require('@librechat/api');
|
||||
const {
|
||||
Tokenizer,
|
||||
createFetch,
|
||||
matchModelName,
|
||||
getClaudeHeaders,
|
||||
getModelMaxTokens,
|
||||
configureReasoning,
|
||||
checkPromptCacheSupport,
|
||||
getModelMaxOutputTokens,
|
||||
createStreamEventHandlers,
|
||||
} = require('@librechat/api');
|
||||
const {
|
||||
truncateText,
|
||||
formatMessage,
|
||||
|
|
@ -19,12 +29,6 @@ const {
|
|||
parseParamFromPrompt,
|
||||
createContextHandlers,
|
||||
} = require('./prompts');
|
||||
const {
|
||||
getClaudeHeaders,
|
||||
configureReasoning,
|
||||
checkPromptCacheSupport,
|
||||
} = require('~/server/services/Endpoints/anthropic/helpers');
|
||||
const { getModelMaxTokens, getModelMaxOutputTokens, matchModelName } = require('~/utils');
|
||||
const { spendTokens, spendStructuredTokens } = require('~/models/spendTokens');
|
||||
const { encodeAndFormat } = require('~/server/services/Files/images/encode');
|
||||
const { sleep } = require('~/server/utils');
|
||||
|
|
|
|||
|
|
@ -1,4 +1,5 @@
|
|||
const { google } = require('googleapis');
|
||||
const { getModelMaxTokens } = require('@librechat/api');
|
||||
const { concat } = require('@langchain/core/utils/stream');
|
||||
const { ChatVertexAI } = require('@langchain/google-vertexai');
|
||||
const { Tokenizer, getSafetySettings } = require('@librechat/api');
|
||||
|
|
@ -21,7 +22,6 @@ const {
|
|||
} = require('librechat-data-provider');
|
||||
const { encodeAndFormat } = require('~/server/services/Files/images');
|
||||
const { spendTokens } = require('~/models/spendTokens');
|
||||
const { getModelMaxTokens } = require('~/utils');
|
||||
const { sleep } = require('~/server/utils');
|
||||
const { logger } = require('~/config');
|
||||
const {
|
||||
|
|
|
|||
|
|
@ -7,7 +7,9 @@ const {
|
|||
createFetch,
|
||||
resolveHeaders,
|
||||
constructAzureURL,
|
||||
getModelMaxTokens,
|
||||
genAzureChatCompletion,
|
||||
getModelMaxOutputTokens,
|
||||
createStreamEventHandlers,
|
||||
} = require('@librechat/api');
|
||||
const {
|
||||
|
|
@ -31,13 +33,13 @@ const {
|
|||
titleInstruction,
|
||||
createContextHandlers,
|
||||
} = require('./prompts');
|
||||
const { extractBaseURL, getModelMaxTokens, getModelMaxOutputTokens } = require('~/utils');
|
||||
const { encodeAndFormat } = require('~/server/services/Files/images/encode');
|
||||
const { addSpaceIfNeeded, sleep } = require('~/server/utils');
|
||||
const { spendTokens } = require('~/models/spendTokens');
|
||||
const { handleOpenAIErrors } = require('./tools/util');
|
||||
const { summaryBuffer } = require('./memory');
|
||||
const { runTitleChain } = require('./chains');
|
||||
const { extractBaseURL } = require('~/utils');
|
||||
const { tokenSplit } = require('./document');
|
||||
const BaseClient = require('./BaseClient');
|
||||
const { createLLM } = require('./llm');
|
||||
|
|
|
|||
|
|
@ -1,5 +1,5 @@
|
|||
const { getModelMaxTokens } = require('@librechat/api');
|
||||
const BaseClient = require('../BaseClient');
|
||||
const { getModelMaxTokens } = require('../../../utils');
|
||||
|
||||
class FakeClient extends BaseClient {
|
||||
constructor(apiKey, options = {}) {
|
||||
|
|
|
|||
|
|
@ -1,4 +1,4 @@
|
|||
const { matchModelName } = require('../utils/tokens');
|
||||
const { matchModelName } = require('@librechat/api');
|
||||
const defaultRate = 6;
|
||||
|
||||
/**
|
||||
|
|
|
|||
|
|
@ -49,7 +49,7 @@
|
|||
"@langchain/google-vertexai": "^0.2.13",
|
||||
"@langchain/openai": "^0.5.18",
|
||||
"@langchain/textsplitters": "^0.1.0",
|
||||
"@librechat/agents": "^2.4.77",
|
||||
"@librechat/agents": "^2.4.79",
|
||||
"@librechat/api": "*",
|
||||
"@librechat/data-schemas": "*",
|
||||
"@microsoft/microsoft-graph-client": "^3.0.7",
|
||||
|
|
|
|||
|
|
@ -872,11 +872,10 @@ class AgentClient extends BaseClient {
|
|||
if (agent.useLegacyContent === true) {
|
||||
messages = formatContentStrings(messages);
|
||||
}
|
||||
if (
|
||||
agent.model_parameters?.clientOptions?.defaultHeaders?.['anthropic-beta']?.includes(
|
||||
'prompt-caching',
|
||||
)
|
||||
) {
|
||||
const defaultHeaders =
|
||||
agent.model_parameters?.clientOptions?.defaultHeaders ??
|
||||
agent.model_parameters?.configuration?.defaultHeaders;
|
||||
if (defaultHeaders?.['anthropic-beta']?.includes('prompt-caching')) {
|
||||
messages = addCacheControl(messages);
|
||||
}
|
||||
|
||||
|
|
|
|||
|
|
@ -1,7 +1,7 @@
|
|||
const { v4 } = require('uuid');
|
||||
const { sleep } = require('@librechat/agents');
|
||||
const { logger } = require('@librechat/data-schemas');
|
||||
const { sendEvent, getBalanceConfig } = require('@librechat/api');
|
||||
const { sendEvent, getBalanceConfig, getModelMaxTokens } = require('@librechat/api');
|
||||
const {
|
||||
Time,
|
||||
Constants,
|
||||
|
|
@ -34,7 +34,6 @@ const { checkBalance } = require('~/models/balanceMethods');
|
|||
const { getConvo } = require('~/models/Conversation');
|
||||
const getLogStores = require('~/cache/getLogStores');
|
||||
const { countTokens } = require('~/server/utils');
|
||||
const { getModelMaxTokens } = require('~/utils');
|
||||
const { getOpenAIClient } = require('./helpers');
|
||||
|
||||
/**
|
||||
|
|
|
|||
|
|
@ -1,7 +1,7 @@
|
|||
const { v4 } = require('uuid');
|
||||
const { sleep } = require('@librechat/agents');
|
||||
const { logger } = require('@librechat/data-schemas');
|
||||
const { sendEvent, getBalanceConfig } = require('@librechat/api');
|
||||
const { sendEvent, getBalanceConfig, getModelMaxTokens } = require('@librechat/api');
|
||||
const {
|
||||
Time,
|
||||
Constants,
|
||||
|
|
@ -31,7 +31,6 @@ const { checkBalance } = require('~/models/balanceMethods');
|
|||
const { getConvo } = require('~/models/Conversation');
|
||||
const getLogStores = require('~/cache/getLogStores');
|
||||
const { countTokens } = require('~/server/utils');
|
||||
const { getModelMaxTokens } = require('~/utils');
|
||||
const { getOpenAIClient } = require('./helpers');
|
||||
|
||||
/**
|
||||
|
|
|
|||
|
|
@ -1,6 +1,7 @@
|
|||
const { Providers } = require('@librechat/agents');
|
||||
const {
|
||||
primeResources,
|
||||
getModelMaxTokens,
|
||||
extractLibreChatParams,
|
||||
optionalChainWithEmptyCheck,
|
||||
} = require('@librechat/api');
|
||||
|
|
@ -17,7 +18,6 @@ const { getProviderConfig } = require('~/server/services/Endpoints');
|
|||
const { processFiles } = require('~/server/services/Files/process');
|
||||
const { getFiles, getToolFilesByIds } = require('~/models/File');
|
||||
const { getConvoFiles } = require('~/models/Conversation');
|
||||
const { getModelMaxTokens } = require('~/utils');
|
||||
|
||||
/**
|
||||
* @param {object} params
|
||||
|
|
|
|||
|
|
@ -1,118 +0,0 @@
|
|||
const { EModelEndpoint, anthropicSettings } = require('librechat-data-provider');
|
||||
const { matchModelName } = require('~/utils');
|
||||
const { logger } = require('~/config');
|
||||
|
||||
/**
|
||||
* @param {string} modelName
|
||||
* @returns {boolean}
|
||||
*/
|
||||
function checkPromptCacheSupport(modelName) {
|
||||
const modelMatch = matchModelName(modelName, EModelEndpoint.anthropic);
|
||||
if (
|
||||
modelMatch.includes('claude-3-5-sonnet-latest') ||
|
||||
modelMatch.includes('claude-3.5-sonnet-latest')
|
||||
) {
|
||||
return false;
|
||||
}
|
||||
|
||||
return (
|
||||
/claude-3[-.]7/.test(modelMatch) ||
|
||||
/claude-3[-.]5-(?:sonnet|haiku)/.test(modelMatch) ||
|
||||
/claude-3-(?:sonnet|haiku|opus)?/.test(modelMatch) ||
|
||||
/claude-(?:sonnet|opus|haiku)-[4-9]/.test(modelMatch) ||
|
||||
/claude-[4-9]-(?:sonnet|opus|haiku)?/.test(modelMatch) ||
|
||||
/claude-4(?:-(?:sonnet|opus|haiku))?/.test(modelMatch)
|
||||
);
|
||||
}
|
||||
|
||||
/**
|
||||
* Gets the appropriate headers for Claude models with cache control
|
||||
* @param {string} model The model name
|
||||
* @param {boolean} supportsCacheControl Whether the model supports cache control
|
||||
* @returns {AnthropicClientOptions['extendedOptions']['defaultHeaders']|undefined} The headers object or undefined if not applicable
|
||||
*/
|
||||
function getClaudeHeaders(model, supportsCacheControl) {
|
||||
if (!supportsCacheControl) {
|
||||
return undefined;
|
||||
}
|
||||
|
||||
if (/claude-3[-.]5-sonnet/.test(model)) {
|
||||
return {
|
||||
'anthropic-beta': 'max-tokens-3-5-sonnet-2024-07-15,prompt-caching-2024-07-31',
|
||||
};
|
||||
} else if (/claude-3[-.]7/.test(model)) {
|
||||
return {
|
||||
'anthropic-beta':
|
||||
'token-efficient-tools-2025-02-19,output-128k-2025-02-19,prompt-caching-2024-07-31',
|
||||
};
|
||||
} else if (/claude-sonnet-4/.test(model)) {
|
||||
return {
|
||||
'anthropic-beta': 'prompt-caching-2024-07-31,context-1m-2025-08-07',
|
||||
};
|
||||
} else if (
|
||||
/claude-(?:sonnet|opus|haiku)-[4-9]/.test(model) ||
|
||||
/claude-[4-9]-(?:sonnet|opus|haiku)?/.test(model) ||
|
||||
/claude-4(?:-(?:sonnet|opus|haiku))?/.test(model)
|
||||
) {
|
||||
return {
|
||||
'anthropic-beta': 'prompt-caching-2024-07-31',
|
||||
};
|
||||
} else {
|
||||
return {
|
||||
'anthropic-beta': 'prompt-caching-2024-07-31',
|
||||
};
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* Configures reasoning-related options for Claude models
|
||||
* @param {AnthropicClientOptions & { max_tokens?: number }} anthropicInput The request options object
|
||||
* @param {Object} extendedOptions Additional client configuration options
|
||||
* @param {boolean} extendedOptions.thinking Whether thinking is enabled in client config
|
||||
* @param {number|null} extendedOptions.thinkingBudget The token budget for thinking
|
||||
* @returns {Object} Updated request options
|
||||
*/
|
||||
function configureReasoning(anthropicInput, extendedOptions = {}) {
|
||||
const updatedOptions = { ...anthropicInput };
|
||||
const currentMaxTokens = updatedOptions.max_tokens ?? updatedOptions.maxTokens;
|
||||
if (
|
||||
extendedOptions.thinking &&
|
||||
updatedOptions?.model &&
|
||||
(/claude-3[-.]7/.test(updatedOptions.model) ||
|
||||
/claude-(?:sonnet|opus|haiku)-[4-9]/.test(updatedOptions.model))
|
||||
) {
|
||||
updatedOptions.thinking = {
|
||||
type: 'enabled',
|
||||
};
|
||||
}
|
||||
|
||||
if (updatedOptions.thinking != null && extendedOptions.thinkingBudget != null) {
|
||||
updatedOptions.thinking = {
|
||||
...updatedOptions.thinking,
|
||||
budget_tokens: extendedOptions.thinkingBudget,
|
||||
};
|
||||
}
|
||||
|
||||
if (
|
||||
updatedOptions.thinking != null &&
|
||||
(currentMaxTokens == null || updatedOptions.thinking.budget_tokens > currentMaxTokens)
|
||||
) {
|
||||
const maxTokens = anthropicSettings.maxOutputTokens.reset(updatedOptions.model);
|
||||
updatedOptions.max_tokens = currentMaxTokens ?? maxTokens;
|
||||
|
||||
logger.warn(
|
||||
updatedOptions.max_tokens === maxTokens
|
||||
? '[AnthropicClient] max_tokens is not defined while thinking is enabled. Setting max_tokens to model default.'
|
||||
: `[AnthropicClient] thinking budget_tokens (${updatedOptions.thinking.budget_tokens}) exceeds max_tokens (${updatedOptions.max_tokens}). Adjusting budget_tokens.`,
|
||||
);
|
||||
|
||||
updatedOptions.thinking.budget_tokens = Math.min(
|
||||
updatedOptions.thinking.budget_tokens,
|
||||
Math.floor(updatedOptions.max_tokens * 0.9),
|
||||
);
|
||||
}
|
||||
|
||||
return updatedOptions;
|
||||
}
|
||||
|
||||
module.exports = { checkPromptCacheSupport, getClaudeHeaders, configureReasoning };
|
||||
|
|
@ -1,6 +1,6 @@
|
|||
const { getLLMConfig } = require('@librechat/api');
|
||||
const { EModelEndpoint } = require('librechat-data-provider');
|
||||
const { getUserKey, checkUserKeyExpiry } = require('~/server/services/UserService');
|
||||
const { getLLMConfig } = require('~/server/services/Endpoints/anthropic/llm');
|
||||
const AnthropicClient = require('~/app/clients/AnthropicClient');
|
||||
|
||||
const initializeClient = async ({ req, res, endpointOption, overrideModel, optionsOnly }) => {
|
||||
|
|
@ -40,7 +40,6 @@ const initializeClient = async ({ req, res, endpointOption, overrideModel, optio
|
|||
clientOptions = Object.assign(
|
||||
{
|
||||
proxy: PROXY ?? null,
|
||||
userId: req.user.id,
|
||||
reverseProxyUrl: ANTHROPIC_REVERSE_PROXY ?? null,
|
||||
modelOptions: endpointOption?.model_parameters ?? {},
|
||||
},
|
||||
|
|
@ -49,6 +48,7 @@ const initializeClient = async ({ req, res, endpointOption, overrideModel, optio
|
|||
if (overrideModel) {
|
||||
clientOptions.modelOptions.model = overrideModel;
|
||||
}
|
||||
clientOptions.modelOptions.user = req.user.id;
|
||||
return getLLMConfig(anthropicApiKey, clientOptions);
|
||||
}
|
||||
|
||||
|
|
|
|||
|
|
@ -1,103 +0,0 @@
|
|||
const { ProxyAgent } = require('undici');
|
||||
const { anthropicSettings, removeNullishValues } = require('librechat-data-provider');
|
||||
const { checkPromptCacheSupport, getClaudeHeaders, configureReasoning } = require('./helpers');
|
||||
|
||||
/**
|
||||
* Generates configuration options for creating an Anthropic language model (LLM) instance.
|
||||
*
|
||||
* @param {string} apiKey - The API key for authentication with Anthropic.
|
||||
* @param {Object} [options={}] - Additional options for configuring the LLM.
|
||||
* @param {Object} [options.modelOptions] - Model-specific options.
|
||||
* @param {string} [options.modelOptions.model] - The name of the model to use.
|
||||
* @param {number} [options.modelOptions.maxOutputTokens] - The maximum number of tokens to generate.
|
||||
* @param {number} [options.modelOptions.temperature] - Controls randomness in output generation.
|
||||
* @param {number} [options.modelOptions.topP] - Controls diversity of output generation.
|
||||
* @param {number} [options.modelOptions.topK] - Controls the number of top tokens to consider.
|
||||
* @param {string[]} [options.modelOptions.stop] - Sequences where the API will stop generating further tokens.
|
||||
* @param {boolean} [options.modelOptions.stream] - Whether to stream the response.
|
||||
* @param {string} options.userId - The user ID for tracking and personalization.
|
||||
* @param {string} [options.proxy] - Proxy server URL.
|
||||
* @param {string} [options.reverseProxyUrl] - URL for a reverse proxy, if used.
|
||||
*
|
||||
* @returns {Object} Configuration options for creating an Anthropic LLM instance, with null and undefined values removed.
|
||||
*/
|
||||
function getLLMConfig(apiKey, options = {}) {
|
||||
const systemOptions = {
|
||||
thinking: options.modelOptions.thinking ?? anthropicSettings.thinking.default,
|
||||
promptCache: options.modelOptions.promptCache ?? anthropicSettings.promptCache.default,
|
||||
thinkingBudget: options.modelOptions.thinkingBudget ?? anthropicSettings.thinkingBudget.default,
|
||||
};
|
||||
for (let key in systemOptions) {
|
||||
delete options.modelOptions[key];
|
||||
}
|
||||
const defaultOptions = {
|
||||
model: anthropicSettings.model.default,
|
||||
maxOutputTokens: anthropicSettings.maxOutputTokens.default,
|
||||
stream: true,
|
||||
};
|
||||
|
||||
const mergedOptions = Object.assign(defaultOptions, options.modelOptions);
|
||||
|
||||
/** @type {AnthropicClientOptions} */
|
||||
let requestOptions = {
|
||||
apiKey,
|
||||
model: mergedOptions.model,
|
||||
stream: mergedOptions.stream,
|
||||
temperature: mergedOptions.temperature,
|
||||
stopSequences: mergedOptions.stop,
|
||||
maxTokens:
|
||||
mergedOptions.maxOutputTokens || anthropicSettings.maxOutputTokens.reset(mergedOptions.model),
|
||||
clientOptions: {},
|
||||
invocationKwargs: {
|
||||
metadata: {
|
||||
user_id: options.userId,
|
||||
},
|
||||
},
|
||||
};
|
||||
|
||||
requestOptions = configureReasoning(requestOptions, systemOptions);
|
||||
|
||||
if (!/claude-3[-.]7/.test(mergedOptions.model)) {
|
||||
requestOptions.topP = mergedOptions.topP;
|
||||
requestOptions.topK = mergedOptions.topK;
|
||||
} else if (requestOptions.thinking == null) {
|
||||
requestOptions.topP = mergedOptions.topP;
|
||||
requestOptions.topK = mergedOptions.topK;
|
||||
}
|
||||
|
||||
const supportsCacheControl =
|
||||
systemOptions.promptCache === true && checkPromptCacheSupport(requestOptions.model);
|
||||
const headers = getClaudeHeaders(requestOptions.model, supportsCacheControl);
|
||||
if (headers) {
|
||||
requestOptions.clientOptions.defaultHeaders = headers;
|
||||
}
|
||||
|
||||
if (options.proxy) {
|
||||
const proxyAgent = new ProxyAgent(options.proxy);
|
||||
requestOptions.clientOptions.fetchOptions = {
|
||||
dispatcher: proxyAgent,
|
||||
};
|
||||
}
|
||||
|
||||
if (options.reverseProxyUrl) {
|
||||
requestOptions.clientOptions.baseURL = options.reverseProxyUrl;
|
||||
requestOptions.anthropicApiUrl = options.reverseProxyUrl;
|
||||
}
|
||||
|
||||
const tools = [];
|
||||
|
||||
if (mergedOptions.web_search) {
|
||||
tools.push({
|
||||
type: 'web_search_20250305',
|
||||
name: 'web_search',
|
||||
});
|
||||
}
|
||||
|
||||
return {
|
||||
tools,
|
||||
/** @type {AnthropicClientOptions} */
|
||||
llmConfig: removeNullishValues(requestOptions),
|
||||
};
|
||||
}
|
||||
|
||||
module.exports = { getLLMConfig };
|
||||
File diff suppressed because it is too large
Load diff
|
|
@ -1,3 +1,4 @@
|
|||
const { getModelMaxTokens } = require('@librechat/api');
|
||||
const { createContentAggregator } = require('@librechat/agents');
|
||||
const {
|
||||
EModelEndpoint,
|
||||
|
|
@ -7,7 +8,6 @@ const {
|
|||
const { getDefaultHandlers } = require('~/server/controllers/agents/callbacks');
|
||||
const getOptions = require('~/server/services/Endpoints/bedrock/options');
|
||||
const AgentClient = require('~/server/controllers/agents/client');
|
||||
const { getModelMaxTokens } = require('~/utils');
|
||||
|
||||
const initializeClient = async ({ req, res, endpointOption }) => {
|
||||
if (!endpointOption) {
|
||||
|
|
|
|||
|
|
@ -1,13 +1,13 @@
|
|||
const axios = require('axios');
|
||||
const { Providers } = require('@librechat/agents');
|
||||
const { logAxiosError } = require('@librechat/api');
|
||||
const { logger } = require('@librechat/data-schemas');
|
||||
const { HttpsProxyAgent } = require('https-proxy-agent');
|
||||
const { logAxiosError, inputSchema, processModelData } = require('@librechat/api');
|
||||
const { EModelEndpoint, defaultModels, CacheKeys } = require('librechat-data-provider');
|
||||
const { inputSchema, extractBaseURL, processModelData } = require('~/utils');
|
||||
const { OllamaClient } = require('~/app/clients/OllamaClient');
|
||||
const { isUserProvided } = require('~/server/utils');
|
||||
const getLogStores = require('~/cache/getLogStores');
|
||||
const { extractBaseURL } = require('~/utils');
|
||||
|
||||
/**
|
||||
* Splits a string by commas and trims each resulting value.
|
||||
|
|
|
|||
|
|
@ -11,8 +11,8 @@ const {
|
|||
getAnthropicModels,
|
||||
} = require('./ModelService');
|
||||
|
||||
jest.mock('~/utils', () => {
|
||||
const originalUtils = jest.requireActual('~/utils');
|
||||
jest.mock('@librechat/api', () => {
|
||||
const originalUtils = jest.requireActual('@librechat/api');
|
||||
return {
|
||||
...originalUtils,
|
||||
processModelData: jest.fn((...args) => {
|
||||
|
|
@ -108,7 +108,7 @@ describe('fetchModels with createTokenConfig true', () => {
|
|||
|
||||
beforeEach(() => {
|
||||
// Clears the mock's history before each test
|
||||
const _utils = require('~/utils');
|
||||
const _utils = require('@librechat/api');
|
||||
axios.get.mockResolvedValue({ data });
|
||||
});
|
||||
|
||||
|
|
@ -120,7 +120,7 @@ describe('fetchModels with createTokenConfig true', () => {
|
|||
createTokenConfig: true,
|
||||
});
|
||||
|
||||
const { processModelData } = require('~/utils');
|
||||
const { processModelData } = require('@librechat/api');
|
||||
expect(processModelData).toHaveBeenCalled();
|
||||
expect(processModelData).toHaveBeenCalledWith(data);
|
||||
});
|
||||
|
|
|
|||
|
|
@ -1,7 +1,7 @@
|
|||
const axios = require('axios');
|
||||
const deriveBaseURL = require('./deriveBaseURL');
|
||||
jest.mock('~/utils', () => {
|
||||
const originalUtils = jest.requireActual('~/utils');
|
||||
jest.mock('@librechat/api', () => {
|
||||
const originalUtils = jest.requireActual('@librechat/api');
|
||||
return {
|
||||
...originalUtils,
|
||||
processModelData: jest.fn((...args) => {
|
||||
|
|
|
|||
|
|
@ -1,4 +1,3 @@
|
|||
const tokenHelpers = require('./tokens');
|
||||
const deriveBaseURL = require('./deriveBaseURL');
|
||||
const extractBaseURL = require('./extractBaseURL');
|
||||
const findMessageContent = require('./findMessageContent');
|
||||
|
|
@ -6,6 +5,5 @@ const findMessageContent = require('./findMessageContent');
|
|||
module.exports = {
|
||||
deriveBaseURL,
|
||||
extractBaseURL,
|
||||
...tokenHelpers,
|
||||
findMessageContent,
|
||||
};
|
||||
|
|
|
|||
|
|
@ -1,493 +0,0 @@
|
|||
const z = require('zod');
|
||||
const { EModelEndpoint } = require('librechat-data-provider');
|
||||
|
||||
const openAIModels = {
|
||||
'o4-mini': 200000,
|
||||
'o3-mini': 195000, // -5000 from max
|
||||
o3: 200000,
|
||||
o1: 195000, // -5000 from max
|
||||
'o1-mini': 127500, // -500 from max
|
||||
'o1-preview': 127500, // -500 from max
|
||||
'gpt-4': 8187, // -5 from max
|
||||
'gpt-4-0613': 8187, // -5 from max
|
||||
'gpt-4-32k': 32758, // -10 from max
|
||||
'gpt-4-32k-0314': 32758, // -10 from max
|
||||
'gpt-4-32k-0613': 32758, // -10 from max
|
||||
'gpt-4-1106': 127500, // -500 from max
|
||||
'gpt-4-0125': 127500, // -500 from max
|
||||
'gpt-4.5': 127500, // -500 from max
|
||||
'gpt-4.1': 1047576,
|
||||
'gpt-4.1-mini': 1047576,
|
||||
'gpt-4.1-nano': 1047576,
|
||||
'gpt-5': 400000,
|
||||
'gpt-5-mini': 400000,
|
||||
'gpt-5-nano': 400000,
|
||||
'gpt-4o': 127500, // -500 from max
|
||||
'gpt-4o-mini': 127500, // -500 from max
|
||||
'gpt-4o-2024-05-13': 127500, // -500 from max
|
||||
'gpt-4o-2024-08-06': 127500, // -500 from max
|
||||
'gpt-4-turbo': 127500, // -500 from max
|
||||
'gpt-4-vision': 127500, // -500 from max
|
||||
'gpt-3.5-turbo': 16375, // -10 from max
|
||||
'gpt-3.5-turbo-0613': 4092, // -5 from max
|
||||
'gpt-3.5-turbo-0301': 4092, // -5 from max
|
||||
'gpt-3.5-turbo-16k': 16375, // -10 from max
|
||||
'gpt-3.5-turbo-16k-0613': 16375, // -10 from max
|
||||
'gpt-3.5-turbo-1106': 16375, // -10 from max
|
||||
'gpt-3.5-turbo-0125': 16375, // -10 from max
|
||||
};
|
||||
|
||||
const mistralModels = {
|
||||
'mistral-': 31990, // -10 from max
|
||||
'mistral-7b': 31990, // -10 from max
|
||||
'mistral-small': 31990, // -10 from max
|
||||
'mixtral-8x7b': 31990, // -10 from max
|
||||
'mistral-large': 131000,
|
||||
'mistral-large-2402': 127500,
|
||||
'mistral-large-2407': 127500,
|
||||
'pixtral-large': 131000,
|
||||
'mistral-saba': 32000,
|
||||
codestral: 256000,
|
||||
'ministral-8b': 131000,
|
||||
'ministral-3b': 131000,
|
||||
};
|
||||
|
||||
const cohereModels = {
|
||||
'command-light': 4086, // -10 from max
|
||||
'command-light-nightly': 8182, // -10 from max
|
||||
command: 4086, // -10 from max
|
||||
'command-nightly': 8182, // -10 from max
|
||||
'command-r': 127500, // -500 from max
|
||||
'command-r-plus': 127500, // -500 from max
|
||||
};
|
||||
|
||||
const googleModels = {
|
||||
/* Max I/O is combined so we subtract the amount from max response tokens for actual total */
|
||||
gemma: 8196,
|
||||
'gemma-2': 32768,
|
||||
'gemma-3': 32768,
|
||||
'gemma-3-27b': 131072,
|
||||
gemini: 30720, // -2048 from max
|
||||
'gemini-pro-vision': 12288,
|
||||
'gemini-exp': 2000000,
|
||||
'gemini-2.5': 1000000, // 1M input tokens, 64k output tokens
|
||||
'gemini-2.5-pro': 1000000,
|
||||
'gemini-2.5-flash': 1000000,
|
||||
'gemini-2.0': 2000000,
|
||||
'gemini-2.0-flash': 1000000,
|
||||
'gemini-2.0-flash-lite': 1000000,
|
||||
'gemini-1.5': 1000000,
|
||||
'gemini-1.5-flash': 1000000,
|
||||
'gemini-1.5-flash-8b': 1000000,
|
||||
'text-bison-32k': 32758, // -10 from max
|
||||
'chat-bison-32k': 32758, // -10 from max
|
||||
'code-bison-32k': 32758, // -10 from max
|
||||
'codechat-bison-32k': 32758,
|
||||
/* Codey, -5 from max: 6144 */
|
||||
'code-': 6139,
|
||||
'codechat-': 6139,
|
||||
/* PaLM2, -5 from max: 8192 */
|
||||
'text-': 8187,
|
||||
'chat-': 8187,
|
||||
};
|
||||
|
||||
const anthropicModels = {
|
||||
'claude-': 100000,
|
||||
'claude-instant': 100000,
|
||||
'claude-2': 100000,
|
||||
'claude-2.1': 200000,
|
||||
'claude-3': 200000,
|
||||
'claude-3-haiku': 200000,
|
||||
'claude-3-sonnet': 200000,
|
||||
'claude-3-opus': 200000,
|
||||
'claude-3.5-haiku': 200000,
|
||||
'claude-3-5-haiku': 200000,
|
||||
'claude-3-5-sonnet': 200000,
|
||||
'claude-3.5-sonnet': 200000,
|
||||
'claude-3-7-sonnet': 200000,
|
||||
'claude-3.7-sonnet': 200000,
|
||||
'claude-3-5-sonnet-latest': 200000,
|
||||
'claude-3.5-sonnet-latest': 200000,
|
||||
'claude-sonnet-4': 1000000,
|
||||
'claude-opus-4': 200000,
|
||||
'claude-4': 200000,
|
||||
};
|
||||
|
||||
const deepseekModels = {
|
||||
'deepseek-reasoner': 63000, // -1000 from max (API)
|
||||
deepseek: 63000, // -1000 from max (API)
|
||||
'deepseek.r1': 127500,
|
||||
};
|
||||
|
||||
const metaModels = {
|
||||
// Basic patterns
|
||||
llama3: 8000,
|
||||
llama2: 4000,
|
||||
'llama-3': 8000,
|
||||
'llama-2': 4000,
|
||||
|
||||
// llama3.x pattern
|
||||
'llama3.1': 127500,
|
||||
'llama3.2': 127500,
|
||||
'llama3.3': 127500,
|
||||
|
||||
// llama3-x pattern
|
||||
'llama3-1': 127500,
|
||||
'llama3-2': 127500,
|
||||
'llama3-3': 127500,
|
||||
|
||||
// llama-3.x pattern
|
||||
'llama-3.1': 127500,
|
||||
'llama-3.2': 127500,
|
||||
'llama-3.3': 127500,
|
||||
|
||||
// llama3.x:Nb pattern
|
||||
'llama3.1:405b': 127500,
|
||||
'llama3.1:70b': 127500,
|
||||
'llama3.1:8b': 127500,
|
||||
'llama3.2:1b': 127500,
|
||||
'llama3.2:3b': 127500,
|
||||
'llama3.2:11b': 127500,
|
||||
'llama3.2:90b': 127500,
|
||||
'llama3.3:70b': 127500,
|
||||
|
||||
// llama3-x-Nb pattern
|
||||
'llama3-1-405b': 127500,
|
||||
'llama3-1-70b': 127500,
|
||||
'llama3-1-8b': 127500,
|
||||
'llama3-2-1b': 127500,
|
||||
'llama3-2-3b': 127500,
|
||||
'llama3-2-11b': 127500,
|
||||
'llama3-2-90b': 127500,
|
||||
'llama3-3-70b': 127500,
|
||||
|
||||
// llama-3.x-Nb pattern
|
||||
'llama-3.1-405b': 127500,
|
||||
'llama-3.1-70b': 127500,
|
||||
'llama-3.1-8b': 127500,
|
||||
'llama-3.2-1b': 127500,
|
||||
'llama-3.2-3b': 127500,
|
||||
'llama-3.2-11b': 127500,
|
||||
'llama-3.2-90b': 127500,
|
||||
'llama-3.3-70b': 127500,
|
||||
|
||||
// Original llama2/3 patterns
|
||||
'llama3-70b': 8000,
|
||||
'llama3-8b': 8000,
|
||||
'llama2-70b': 4000,
|
||||
'llama2-13b': 4000,
|
||||
'llama3:70b': 8000,
|
||||
'llama3:8b': 8000,
|
||||
'llama2:70b': 4000,
|
||||
};
|
||||
|
||||
const ollamaModels = {
|
||||
'qwen2.5': 32000,
|
||||
};
|
||||
|
||||
const ai21Models = {
|
||||
'ai21.j2-mid-v1': 8182, // -10 from max
|
||||
'ai21.j2-ultra-v1': 8182, // -10 from max
|
||||
'ai21.jamba-instruct-v1:0': 255500, // -500 from max
|
||||
};
|
||||
|
||||
const amazonModels = {
|
||||
'amazon.titan-text-lite-v1': 4000,
|
||||
'amazon.titan-text-express-v1': 8000,
|
||||
'amazon.titan-text-premier-v1:0': 31500, // -500 from max
|
||||
// https://aws.amazon.com/ai/generative-ai/nova/
|
||||
'amazon.nova-micro-v1:0': 127000, // -1000 from max,
|
||||
'amazon.nova-lite-v1:0': 295000, // -5000 from max,
|
||||
'amazon.nova-pro-v1:0': 295000, // -5000 from max,
|
||||
'amazon.nova-premier-v1:0': 995000, // -5000 from max,
|
||||
};
|
||||
|
||||
const bedrockModels = {
|
||||
...anthropicModels,
|
||||
...mistralModels,
|
||||
...cohereModels,
|
||||
...ollamaModels,
|
||||
...deepseekModels,
|
||||
...metaModels,
|
||||
...ai21Models,
|
||||
...amazonModels,
|
||||
};
|
||||
|
||||
const xAIModels = {
|
||||
grok: 131072,
|
||||
'grok-beta': 131072,
|
||||
'grok-vision-beta': 8192,
|
||||
'grok-2': 131072,
|
||||
'grok-2-latest': 131072,
|
||||
'grok-2-1212': 131072,
|
||||
'grok-2-vision': 32768,
|
||||
'grok-2-vision-latest': 32768,
|
||||
'grok-2-vision-1212': 32768,
|
||||
'grok-3': 131072,
|
||||
'grok-3-fast': 131072,
|
||||
'grok-3-mini': 131072,
|
||||
'grok-3-mini-fast': 131072,
|
||||
'grok-4': 256000, // 256K context
|
||||
};
|
||||
|
||||
const aggregateModels = {
|
||||
...openAIModels,
|
||||
...googleModels,
|
||||
...bedrockModels,
|
||||
...xAIModels,
|
||||
// misc.
|
||||
kimi: 131000,
|
||||
// GPT-OSS
|
||||
'gpt-oss-20b': 131000,
|
||||
'gpt-oss-120b': 131000,
|
||||
};
|
||||
|
||||
const maxTokensMap = {
|
||||
[EModelEndpoint.azureOpenAI]: openAIModels,
|
||||
[EModelEndpoint.openAI]: aggregateModels,
|
||||
[EModelEndpoint.agents]: aggregateModels,
|
||||
[EModelEndpoint.custom]: aggregateModels,
|
||||
[EModelEndpoint.google]: googleModels,
|
||||
[EModelEndpoint.anthropic]: anthropicModels,
|
||||
[EModelEndpoint.bedrock]: bedrockModels,
|
||||
};
|
||||
|
||||
const modelMaxOutputs = {
|
||||
o1: 32268, // -500 from max: 32,768
|
||||
'o1-mini': 65136, // -500 from max: 65,536
|
||||
'o1-preview': 32268, // -500 from max: 32,768
|
||||
'gpt-5': 128000,
|
||||
'gpt-5-mini': 128000,
|
||||
'gpt-5-nano': 128000,
|
||||
'gpt-oss-20b': 131000,
|
||||
'gpt-oss-120b': 131000,
|
||||
system_default: 1024,
|
||||
};
|
||||
|
||||
/** Outputs from https://docs.anthropic.com/en/docs/about-claude/models/all-models#model-names */
|
||||
const anthropicMaxOutputs = {
|
||||
'claude-3-haiku': 4096,
|
||||
'claude-3-sonnet': 4096,
|
||||
'claude-3-opus': 4096,
|
||||
'claude-opus-4': 32000,
|
||||
'claude-sonnet-4': 64000,
|
||||
'claude-3.5-sonnet': 8192,
|
||||
'claude-3-5-sonnet': 8192,
|
||||
'claude-3.7-sonnet': 128000,
|
||||
'claude-3-7-sonnet': 128000,
|
||||
};
|
||||
|
||||
const maxOutputTokensMap = {
|
||||
[EModelEndpoint.anthropic]: anthropicMaxOutputs,
|
||||
[EModelEndpoint.azureOpenAI]: modelMaxOutputs,
|
||||
[EModelEndpoint.openAI]: modelMaxOutputs,
|
||||
[EModelEndpoint.custom]: modelMaxOutputs,
|
||||
};
|
||||
|
||||
/**
|
||||
* Finds the first matching pattern in the tokens map.
|
||||
* @param {string} modelName
|
||||
* @param {Record<string, number>} tokensMap
|
||||
* @returns {string|null}
|
||||
*/
|
||||
function findMatchingPattern(modelName, tokensMap) {
|
||||
const keys = Object.keys(tokensMap);
|
||||
for (let i = keys.length - 1; i >= 0; i--) {
|
||||
const modelKey = keys[i];
|
||||
if (modelName.includes(modelKey)) {
|
||||
return modelKey;
|
||||
}
|
||||
}
|
||||
|
||||
return null;
|
||||
}
|
||||
|
||||
/**
|
||||
* Retrieves a token value for a given model name from a tokens map.
|
||||
*
|
||||
* @param {string} modelName - The name of the model to look up.
|
||||
* @param {EndpointTokenConfig | Record<string, number>} tokensMap - The map of model names to token values.
|
||||
* @param {string} [key='context'] - The key to look up in the tokens map.
|
||||
* @returns {number|undefined} The token value for the given model or undefined if no match is found.
|
||||
*/
|
||||
function getModelTokenValue(modelName, tokensMap, key = 'context') {
|
||||
if (typeof modelName !== 'string' || !tokensMap) {
|
||||
return undefined;
|
||||
}
|
||||
|
||||
if (tokensMap[modelName]?.context) {
|
||||
return tokensMap[modelName].context;
|
||||
}
|
||||
|
||||
if (tokensMap[modelName]) {
|
||||
return tokensMap[modelName];
|
||||
}
|
||||
|
||||
const matchedPattern = findMatchingPattern(modelName, tokensMap);
|
||||
|
||||
if (matchedPattern) {
|
||||
const result = tokensMap[matchedPattern];
|
||||
return result?.[key] ?? result ?? tokensMap.system_default;
|
||||
}
|
||||
|
||||
return tokensMap.system_default;
|
||||
}
|
||||
|
||||
/**
|
||||
* Retrieves the maximum tokens for a given model name.
|
||||
*
|
||||
* @param {string} modelName - The name of the model to look up.
|
||||
* @param {string} endpoint - The endpoint (default is 'openAI').
|
||||
* @param {EndpointTokenConfig} [endpointTokenConfig] - Token Config for current endpoint to use for max tokens lookup
|
||||
* @returns {number|undefined} The maximum tokens for the given model or undefined if no match is found.
|
||||
*/
|
||||
function getModelMaxTokens(modelName, endpoint = EModelEndpoint.openAI, endpointTokenConfig) {
|
||||
const tokensMap = endpointTokenConfig ?? maxTokensMap[endpoint];
|
||||
return getModelTokenValue(modelName, tokensMap);
|
||||
}
|
||||
|
||||
/**
|
||||
* Retrieves the maximum output tokens for a given model name.
|
||||
*
|
||||
* @param {string} modelName - The name of the model to look up.
|
||||
* @param {string} endpoint - The endpoint (default is 'openAI').
|
||||
* @param {EndpointTokenConfig} [endpointTokenConfig] - Token Config for current endpoint to use for max tokens lookup
|
||||
* @returns {number|undefined} The maximum output tokens for the given model or undefined if no match is found.
|
||||
*/
|
||||
function getModelMaxOutputTokens(modelName, endpoint = EModelEndpoint.openAI, endpointTokenConfig) {
|
||||
const tokensMap = endpointTokenConfig ?? maxOutputTokensMap[endpoint];
|
||||
return getModelTokenValue(modelName, tokensMap, 'output');
|
||||
}
|
||||
|
||||
/**
|
||||
* Retrieves the model name key for a given model name input. If the exact model name isn't found,
|
||||
* it searches for partial matches within the model name, checking keys in reverse order.
|
||||
*
|
||||
* @param {string} modelName - The name of the model to look up.
|
||||
* @param {string} endpoint - The endpoint (default is 'openAI').
|
||||
* @returns {string|undefined} The model name key for the given model; returns input if no match is found and is string.
|
||||
*
|
||||
* @example
|
||||
* matchModelName('gpt-4-32k-0613'); // Returns 'gpt-4-32k-0613'
|
||||
* matchModelName('gpt-4-32k-unknown'); // Returns 'gpt-4-32k'
|
||||
* matchModelName('unknown-model'); // Returns undefined
|
||||
*/
|
||||
function matchModelName(modelName, endpoint = EModelEndpoint.openAI) {
|
||||
if (typeof modelName !== 'string') {
|
||||
return undefined;
|
||||
}
|
||||
|
||||
const tokensMap = maxTokensMap[endpoint];
|
||||
if (!tokensMap) {
|
||||
return modelName;
|
||||
}
|
||||
|
||||
if (tokensMap[modelName]) {
|
||||
return modelName;
|
||||
}
|
||||
|
||||
const matchedPattern = findMatchingPattern(modelName, tokensMap);
|
||||
return matchedPattern || modelName;
|
||||
}
|
||||
|
||||
const modelSchema = z.object({
|
||||
id: z.string(),
|
||||
pricing: z.object({
|
||||
prompt: z.string(),
|
||||
completion: z.string(),
|
||||
}),
|
||||
context_length: z.number(),
|
||||
});
|
||||
|
||||
const inputSchema = z.object({
|
||||
data: z.array(modelSchema),
|
||||
});
|
||||
|
||||
/**
|
||||
* Processes a list of model data from an API and organizes it into structured data based on URL and specifics of rates and context.
|
||||
* @param {{ data: Array<z.infer<typeof modelSchema>> }} input The input object containing base URL and data fetched from the API.
|
||||
* @returns {EndpointTokenConfig} The processed model data.
|
||||
*/
|
||||
function processModelData(input) {
|
||||
const validationResult = inputSchema.safeParse(input);
|
||||
if (!validationResult.success) {
|
||||
throw new Error('Invalid input data');
|
||||
}
|
||||
const { data } = validationResult.data;
|
||||
|
||||
/** @type {EndpointTokenConfig} */
|
||||
const tokenConfig = {};
|
||||
|
||||
for (const model of data) {
|
||||
const modelKey = model.id;
|
||||
if (modelKey === 'openrouter/auto') {
|
||||
model.pricing = {
|
||||
prompt: '0.00001',
|
||||
completion: '0.00003',
|
||||
};
|
||||
}
|
||||
const prompt = parseFloat(model.pricing.prompt) * 1000000;
|
||||
const completion = parseFloat(model.pricing.completion) * 1000000;
|
||||
|
||||
tokenConfig[modelKey] = {
|
||||
prompt,
|
||||
completion,
|
||||
context: model.context_length,
|
||||
};
|
||||
}
|
||||
|
||||
return tokenConfig;
|
||||
}
|
||||
|
||||
const tiktokenModels = new Set([
|
||||
'text-davinci-003',
|
||||
'text-davinci-002',
|
||||
'text-davinci-001',
|
||||
'text-curie-001',
|
||||
'text-babbage-001',
|
||||
'text-ada-001',
|
||||
'davinci',
|
||||
'curie',
|
||||
'babbage',
|
||||
'ada',
|
||||
'code-davinci-002',
|
||||
'code-davinci-001',
|
||||
'code-cushman-002',
|
||||
'code-cushman-001',
|
||||
'davinci-codex',
|
||||
'cushman-codex',
|
||||
'text-davinci-edit-001',
|
||||
'code-davinci-edit-001',
|
||||
'text-embedding-ada-002',
|
||||
'text-similarity-davinci-001',
|
||||
'text-similarity-curie-001',
|
||||
'text-similarity-babbage-001',
|
||||
'text-similarity-ada-001',
|
||||
'text-search-davinci-doc-001',
|
||||
'text-search-curie-doc-001',
|
||||
'text-search-babbage-doc-001',
|
||||
'text-search-ada-doc-001',
|
||||
'code-search-babbage-code-001',
|
||||
'code-search-ada-code-001',
|
||||
'gpt2',
|
||||
'gpt-4',
|
||||
'gpt-4-0314',
|
||||
'gpt-4-32k',
|
||||
'gpt-4-32k-0314',
|
||||
'gpt-3.5-turbo',
|
||||
'gpt-3.5-turbo-0301',
|
||||
]);
|
||||
|
||||
module.exports = {
|
||||
inputSchema,
|
||||
modelSchema,
|
||||
maxTokensMap,
|
||||
tiktokenModels,
|
||||
maxOutputTokensMap,
|
||||
matchModelName,
|
||||
processModelData,
|
||||
getModelMaxTokens,
|
||||
getModelTokenValue,
|
||||
findMatchingPattern,
|
||||
getModelMaxOutputTokens,
|
||||
};
|
||||
|
|
@ -1,12 +1,12 @@
|
|||
const { EModelEndpoint } = require('librechat-data-provider');
|
||||
const {
|
||||
maxTokensMap,
|
||||
matchModelName,
|
||||
processModelData,
|
||||
getModelMaxTokens,
|
||||
maxOutputTokensMap,
|
||||
findMatchingPattern,
|
||||
getModelMaxTokens,
|
||||
processModelData,
|
||||
matchModelName,
|
||||
maxTokensMap,
|
||||
} = require('./tokens');
|
||||
} = require('@librechat/api');
|
||||
|
||||
describe('getModelMaxTokens', () => {
|
||||
test('should return correct tokens for exact match', () => {
|
||||
|
|
@ -394,7 +394,7 @@ describe('getModelMaxTokens', () => {
|
|||
});
|
||||
|
||||
test('should return correct max output tokens for GPT-5 models', () => {
|
||||
const { getModelMaxOutputTokens } = require('./tokens');
|
||||
const { getModelMaxOutputTokens } = require('@librechat/api');
|
||||
['gpt-5', 'gpt-5-mini', 'gpt-5-nano'].forEach((model) => {
|
||||
expect(getModelMaxOutputTokens(model)).toBe(maxOutputTokensMap[EModelEndpoint.openAI][model]);
|
||||
expect(getModelMaxOutputTokens(model, EModelEndpoint.openAI)).toBe(
|
||||
|
|
@ -407,7 +407,7 @@ describe('getModelMaxTokens', () => {
|
|||
});
|
||||
|
||||
test('should return correct max output tokens for GPT-OSS models', () => {
|
||||
const { getModelMaxOutputTokens } = require('./tokens');
|
||||
const { getModelMaxOutputTokens } = require('@librechat/api');
|
||||
['gpt-oss-20b', 'gpt-oss-120b'].forEach((model) => {
|
||||
expect(getModelMaxOutputTokens(model)).toBe(maxOutputTokensMap[EModelEndpoint.openAI][model]);
|
||||
expect(getModelMaxOutputTokens(model, EModelEndpoint.openAI)).toBe(
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue