Skip to content

Enhanced ChatGPT Clone: Features Agents, MCP, Skills, DeepSeek, Anthropic, AWS, OpenAI, Responses API, Azure, Groq, o1, GPT-5, Mistral, OpenRouter, Vertex AI, Gemini, Artifacts, AI model switching, message search, Code Interpreter, langchain, DALL-E-3, OpenAPI Actions, Functions, Secure Multi-User Auth, Presets, open-source for self-hosting. Active

🦙 feat: Add LLama 3 System Context Length (#2938)

Danny Avila8939d8af379b
diff --git a/api/utils/tokens.js b/api/utils/tokens.js
index b0d9098821..3b973c40cf 100644
--- a/api/utils/tokens.js
+++ b/api/utils/tokens.js
@@ -59,6 +59,8 @@ const openAIModels = {
'gpt-3.5-turbo-1106': 16375, // -10 from max
'gpt-3.5-turbo-0125': 16375, // -10 from max
'mistral-': 31990, // -10 from max
+ llama3: 8187, // -5 from max
+ 'llama-3': 8187, // -5 from max
};
const cohereModels = {
diff --git a/api/utils/tokens.spec.js b/api/utils/tokens.spec.js
index c51610c917..228e298605 100644
--- a/api/utils/tokens.spec.js
+++ b/api/utils/tokens.spec.js
@@ -20,6 +20,18 @@ describe('getModelMaxTokens', () => {
);
});
+ test('should return correct tokens for LLama 3 models', () => {
+ expect(getModelMaxTokens('meta-llama/llama-3-8b')).toBe(
+ maxTokensMap[EModelEndpoint.openAI]['llama-3'],
+ );
+ expect(getModelMaxTokens('meta-llama/llama-3-8b')).toBe(
+ maxTokensMap[EModelEndpoint.openAI]['llama3'],
+ );
+ expect(getModelMaxTokens('llama-3-500b')).toBe(maxTokensMap[EModelEndpoint.openAI]['llama-3']);
+ expect(getModelMaxTokens('llama3-70b')).toBe(maxTokensMap[EModelEndpoint.openAI]['llama3']);
+ expect(getModelMaxTokens('llama3:latest')).toBe(maxTokensMap[EModelEndpoint.openAI]['llama3']);
+ });
+
test('should return undefined for no match', () => {
expect(getModelMaxTokens('unknown-model')).toBeUndefined();
});