Add support for GPT-4o

[ahm750]

Checked other resources

• I added a very descriptive title to this issue.
• I searched the LangChain.js documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.
• I am sure that this is a bug in LangChain.js rather than my code.
• The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).

Example Code

Error Message and Stack Trace (if applicable)

Failed to calculate number of tokens, falling back to approximate count Error: Unknown model
at getEncodingNameForModel (file:///node_modules/js-tiktoken/dist/chunk-HXW345QN.js:224:13)
at encodingForModel (file:///node_modules/@langchain/core/dist/utils/tiktoken.js:19:24)
at ChatOpenAI.getNumTokens (file:///node_modules/@langchain/core/dist/language_models/base.js:174:40)
at file:///node_modules/@langchain/openai/dist/chat_models.js:642:42
at Array.map ()
at ChatOpenAI.getNumTokensFromMessages (file:///node_modules/@langchain/openai/dist/chat_models.js:641:60)
at ChatOpenAI.getEstimatedTokenCountFromPrompt (file:///node_modules/@langchain/openai/dist/chat_models.js:589:34)
at ChatOpenAI._generate (file:///node_modules/@langchain/openai/dist/chat_models.js:536:49)
at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
at async Promise.allSettled (index 0)
Failed to calculate number of tokens, falling back to approximate count Error: Unknown model
at getEncodingNameForModel (file:///node_modules/js-tiktoken/dist/chunk-HXW345QN.js:224:13)
at encodingForModel (file:///node_modules/@langchain/core/dist/utils/tiktoken.js:19:24)
at ChatOpenAI.getNumTokens (file:///node_modules/@langchain/core/dist/language_models/base.js:174:40)
at file:///node_modules/@langchain/openai/dist/chat_models.js:643:42
at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
at async Promise.all (index 0)
at async ChatOpenAI.getNumTokensFromMessages (file:///node_modules/@langchain/openai/dist/chat_models.js:641:33)
at async ChatOpenAI.getEstimatedTokenCountFromPrompt (file:///node_modules/@langchain/openai/dist/chat_models.js:589:23)
at async ChatOpenAI._generate (file:///node_modules/@langchain/openai/dist/chat_models.js:536:38)
at async Promise.allSettled (index 0)
Failed to calculate number of tokens, falling back to approximate count Error: Unknown model
at getEncodingNameForModel (file:///node_modules/js-tiktoken/dist/chunk-HXW345QN.js:224:13)
at encodingForModel (file:///node_modules/@langchain/core/dist/utils/tiktoken.js:19:24)
at ChatOpenAI.getNumTokens (file:///node_modules/@langchain/core/dist/language_models/base.js:174:40)
at file:///node_modules/@langchain/openai/dist/chat_models.js:623:35
at Array.map ()
at ChatOpenAI.getNumTokensFromGenerations (file:///node_modules/@langchain/openai/dist/chat_models.js:617:64)
at ChatOpenAI._generate (file:///node_modules/@langchain/openai/dist/chat_models.js:537:53)
at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
at async Promise.allSettled (index 0)
at async ChatOpenAI._generateUncached (file:///node_modules/@langchain/core/dist/language_models/chat_models.js:114:25)

Description

Hi, OpenAI just released GPT-4o and langchain needs to be updated to support this model. At the moment, it throws the above error when using the new model.

System Info

Langchain: 0.1.37
Platform: macos
Node: 21.7.3
Yarn: 4.2.2

[jeasonnow]

#5354
This is a resolved issue, please wait for the new version of langchainjs.

[sanjiv-upgiant]

This is not yet solved. Here's my langchain packages.

    "@langchain/community": "^0.0.54",
    "@langchain/groq": "^0.0.9",
    "@langchain/langgraph": "^0.0.12",
    "@langchain/openai": "^0.0.29",

[sanjiv-upgiant]

NEED THE UPDATE TO BE RELEASED........

[ahm750]

@jacoblee93 Any ETA on the new release?

[jacoblee93]

Hey the fix should be live if you update to the latest @langchain/core! The tiktoken dep is there.

[jacoblee93]

https://x.com/philippec_me/status/1790142477141147884?s=46&t=pBOviutddr6W7u_gV0VQqg

[maikosantos]

Hi,

Now, It's working for me!

Thanks!

[ahm750]

Thank you, it works now!