Andreas/LibreChat
1
0
Fork 0
mirror of https://github.com/danny-avila/LibreChat.git synced 2025-12-17 00:40:14 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 6

🚀 feat: o1 (#4019)

Browse source 
* feat: o1 default response sender string

* feat: add o1 models to default openai models list, add `no_system_messages` error type; refactor: use error type as localization key

* refactor(MessageEndpointIcon): differentiate openAI icon model color for o1 models

* refactor(AnthropicClient): use new input/output tokens keys; add prompt caching for claude-3-opus

* refactor(BaseClient): to use new input/output tokens keys; update typedefs

* feat: initial o1 model handling, including token cost complexity

* EXPERIMENTAL: special handling for o1 model with custom instructions
This commit is contained in:
Danny Avila 2024-09-12 18:15:43 -04:00 • committed by GitHub
parent 9a393be012
commit 45b42830a5
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
12 changed files with 229 additions and 36 deletions
Download patch file Download diff file Expand all files Collapse all files

16
api/app/clients/AnthropicClient.js
View file

@ -64,6 +64,12 @@ class AnthropicClient extends BaseClient {
/** Whether or not the model supports Prompt Caching
* @type {boolean} */
this.supportsCacheControl;
/** The key for the usage object's input tokens
* @type {string} */
this.inputTokensKey = 'input_tokens';
/** The key for the usage object's output tokens
* @type {string} */
this.outputTokensKey = 'output_tokens';
}
setOptions(options) {
@ -200,7 +206,7 @@ class AnthropicClient extends BaseClient {
}
/**
* Calculates the correct token count for the current message based on the token count map and API usage.
* Calculates the correct token count for the current user message based on the token count map and API usage.
* Edge case: If the calculation results in a negative value, it returns the original estimate.
* If revisiting a conversation with a chat history entirely composed of token estimates,
* the cumulative token count going forward should become more accurate as the conversation progresses.
@ -208,7 +214,7 @@ class AnthropicClient extends BaseClient {
* @param {Record<string, number>} params.tokenCountMap - A map of message IDs to their token counts.
* @param {string} params.currentMessageId - The ID of the current message to calculate.
* @param {AnthropicStreamUsage} params.usage - The usage object returned by the API.
* @returns {number} The correct token count for the current message.
* @returns {number} The correct token count for the current user message.
*/
calculateCurrentTokenCount({ tokenCountMap, currentMessageId, usage }) {
const originalEstimate = tokenCountMap[currentMessageId] || 0;
@ -680,7 +686,11 @@ class AnthropicClient extends BaseClient {
*/
checkPromptCacheSupport(modelName) {
const modelMatch = matchModelName(modelName, EModelEndpoint.anthropic);
if (modelMatch === 'claude-3-5-sonnet' || modelMatch === 'claude-3-haiku') {
if (
modelMatch === 'claude-3-5-sonnet' ||
modelMatch === 'claude-3-haiku' ||
modelMatch === 'claude-3-opus'
) {
return true;
}
return false;