ChatXAI returns malformed usage_metadata

[NicholasDullam]

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

await ChatXAI({ model: 'grok-3-mini', temperature: 0.3, reasoningEffort: 'high' }).invoke(...)

AIMessageChunk {
  ...
  "response_metadata": {
    "usage": {
      "prompt_tokens": 2645,
      "completion_tokens": 30,
      "total_tokens": 4378,
      "prompt_tokens_details": {
        "text_tokens": 2645,
        "audio_tokens": 0,
        "image_tokens": 0,
        "cached_tokens": 0
      },
      "completion_tokens_details": {
        "reasoning_tokens": 1703,
        "audio_tokens": 0,
        "accepted_prediction_tokens": 0,
        "rejected_prediction_tokens": 0
      }
    }
  },
  ...
  "usage_metadata": {
    "input_tokens": null,
    "output_tokens": null,
    "total_tokens": null,
    "input_token_details": {
      "audio": 0,
      "cache_read": 0
    },
    "output_token_details": {
      "audio": 0,
      "reasoning": 1703
    }
  }
}

Error Message and Stack Trace (if applicable)

No response

Description

When using invoke with grok-3-mini or grok-3-mini-fast under ChatXAI, the usage_metadata is not properly extracted from the model API response; the proper, raw response can be viewed under the response_metadata.

The usage_metadata should include the proper token counts; and should propagate properly in the resulting LangSmith dashboards.

System Info

[email protected] | MIT | deps: 12 | versions: 321
Typescript bindings for langchain
https://github.com/langchain-ai/langchainjs/tree/main/langchain/

keywords: llm, ai, gpt3, chain, prompt, prompt engineering, chatgpt, machine learning, ml, openai, embeddings, vectorstores

dist
.tarball: https://registry.npmjs.org/langchain/-/langchain-0.3.26.tgz
.shasum: 5e1dec39172d1589bfee369a710b2e2459c3df8e
.integrity: sha512-W/9phB4wiAnj+PnpMWmv/ptIp7i5ygY2aK8yjKlxccHPbaNeMoy7njzFz8d0/xfcPyA3MvG4AuZnJ1j3/E2/Ig==
.unpackedSize: 2.9 MB

dependencies:
@langchain/openai: >=0.1.0 <0.6.0 jsonpointer: ^5.0.1 uuid: ^10.0.0
@langchain/textsplitters: >=0.0.0 <0.2.0 langsmith: ^0.3.29 yaml: ^2.2.1
js-tiktoken: ^1.0.12 openapi-types: ^12.1.3 zod-to-json-schema: ^3.22.3
js-yaml: ^4.1.0 p-retry: 4 zod: ^3.22.4

maintainers:

• nfcampos [email protected]
• jacoblee93 [email protected]
• andrewnguonly [email protected]
• benjamincburns [email protected]
• davidduong [email protected]
• hwchase17 [email protected]
• basproul [email protected]

dist-tags:
latest: 0.3.26 tag-for-publishing-older-releases: 0.2.20
next: 0.3.2-rc.0

published 5 days ago by benjamincburns [email protected]

node Version: v18.20.8
pnpm version: v10.2.0