11202 - iFlyTek, Spark Embeddings Error

[Syed-Mujeeb]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

import os

os.environ["SPARK_APP_ID"] = "YOUR_APP_ID"
os.environ["SPARK_API_KEY"] = "YOUR_API_KEY"
os.environ["SPARK_API_SECRET"] = "YOUR_API_SECRET"

text_1 = "iFLYTEK is a well-known intelligent speech and artificial intelligence publicly listed company in the Asia-Pacific Region. Since its establishment, the company is devoted to cornerstone technological research in speech and languages, natural language understanding, machine learning, machine reasoning, adaptive learning, and has maintained the world-leading position in those domains. The company actively promotes the development of A.I. products and their sector-based applications, with visions of enabling machines to listen and speak, understand and think, creating a better world with artificial intelligence."
text_2 = "iFLYTEK Open Platform was launched in 2010 by iFLYTEK as China’s first Artificial Intelligence open platform for Mobile Internet and intelligent hardware developers."

query_result = embeddings.embed_query(text_2)
query_result

Description

Request error: 11202, {'header': {'code': 11202, 'message': 'licc failed', 'sid': 'ase000fd997@dx18e35059f4c738d882'}}

SparkLLM Chat works fine for me only embedding is not working - https://python.langchain.com/docs/integrations/chat/sparkllm

System Info

langchain==0.1.12
langchain-community==0.0.28
langchain-core==0.1.31
langchain-openai==0.0.8
langchain-text-splitters==0.0.1

[dosubot[bot]]

Hey there, @Syed-Mujeeb! 😄 Great to see you around again! How's everything going?

The error code 11202 with the message 'licc failed' when using the embed_query function in the LangChain framework with iFlyTek and Spark embeddings, despite the SparkLLM Chat integration working fine, suggests an issue related to licensing or authentication with the iFlyTek's Spark large language model API.

Given the context of the SparkLLM class implementation in the sparkllm.py file, the error could stem from several potential issues:

1. Invalid API Credentials: The app_id, api_key, and api_secret are crucial for authenticating requests to the SparkLLM API. If any of these credentials are incorrect or have not been properly set (either passed as parameters or set as environment variables), the API will reject the request, potentially leading to an error like 11202. Ensure that these credentials are correct and have the necessary permissions.

2. Licensing Restrictions: The error message 'licc failed' might indicate a licensing issue, such as exceeding the usage limits or trying to access features not included in your current subscription plan. Check your iFlyTek developer account to ensure that your license is active and supports the operations you are trying to perform.

3. API Endpoint or Parameters: The default API URL (wss://spark-api.xf-yun.com/v3.1/chat) and domain (generalv3) are set in the _SparkLLMClient initialization. If the endpoint has changed or if the specified domain does not match your licensing terms, this could lead to authentication or licensing errors. Verify that the API endpoint and domain are correct and compatible with your license.

4. Websocket Connection Issues: Since the SparkLLM client uses WebSockets (websocket-client library) to communicate with the API, any issues with establishing or maintaining the WebSocket connection could lead to errors. This includes network issues, firewalls, or proxy settings that block WebSocket connections.

To troubleshoot and resolve the error:

• Double-check your API credentials (app_id, api_key, api_secret) for accuracy.
• Review your iFlyTek account's licensing terms to ensure you have access to the SparkLLM features you are trying to use.
• Confirm that the API endpoint and domain are correct and have not changed.
• Investigate any potential network or WebSocket connection issues that might be interfering with the API communication.

If the issue persists after these checks, consider reaching out to iFlyTek's support for further assistance regarding the specific licensing or authentication issues related to the error code 11202.

Regarding your question about how LangChain handles the authentication process for the iFlyTek and Spark embeddings, it does so through a combination of generating a timestamp, creating a signature using the API key and API secret, and then encoding this signature along with other required headers into an authorization token. This process is encapsulated within the _SparkLLMClient class, specifically in the _create_url method. You can find more details about this process in the sparkllm.py file.

As for the initialization and usage of the Spark embeddings, I wasn't able to find specific information within the repository. However, typically, embeddings are initialized when the model is loaded and then used during the forward pass of the model to generate embeddings for the input data. If you could provide more specific details about what you're looking for, I might be able to give a more precise answer.

Sources

• libs/community/langchain_community/llms/sparkllm.py

About Dosu

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

[Syed-Mujeeb]

I feel, My API Credentials have no issues
Let me check License Restrictions

[morance]

have the same question