What the data for this category looks like
Data sources for this category include the Chinese Center for Disease Control and Prevention Information System, medical institution HIS and infectious disease reporting systems, and laboratory testing systems. The update frequency is daily. Daily documents include infectious disease report cards, epidemic analysis reports, case correction records, clustered epidemic reports, and epidemiological investigation reports. Data fields cover statutory infectious disease classification, case number, diagnosis time, reporting time limit, correction status, and more. The length of individual documents varies widely, ranging from report cards with hundreds of characters to epidemiological analysis materials with thousands of words.
What constraints these characteristics impose on the model access and configuration link
Daily updated data sources require the model access link to support incremental data pulling and incremental summarization, to avoid repeated processing of historical data. Multiple types of document structures require the configuration to adapt to inputs of different text lengths and formats, while also presetting alignment for professional fields related to statutory infectious diseases. Core time-consuming links such as case diagnosis, reporting time limit, and duplicate correction reports need to be verified, so the model's context and output constraints must match the field verification logic of ledger summarization. At the same time, the output results must comply with the standardized format of disease control reports, avoid content that does not meet regulatory requirements, and prevent consequences such as missed or delayed reporting, and delayed epidemic early warnings.
How to set the configurations
Reference only; verify against the target FastGPT release and deployment documentation.
| Configuration Item | Recommended Value Range | Rationale |
|---|---|---|
maxContext | 8000–12000 token | Adapt to the long text input requirements of epidemiological investigation reports and epidemic analysis reports, avoid truncating core monitoring data |
maxTokens | 2000–3000 token | Match the output length of ledger summary and report generation, ensure complete output of analysis conclusions and ledger entries |
temperature | 0.1–0.3 | Reduce random output, ensure the accuracy and format standardization of infectious disease report data |
AIPROXY_API_ENDPOINT | Fill in according to the deployed model service address | Connect to locally or privately deployed large model services, ensure data does not cross compliance boundaries |
AIPROXY_API_TOKEN | Fill in according to the security credential generated by the service | Complete identity verification for the model service, avoid unauthorized access |
toolChoice | auto | Allow the model to independently choose whether to call tools to complete ledger verification and data summarization, adapt to multi-step tasks |
Three Common Misconfigurations
- Symptom: A 401 unauthorized error is returned by the interface when configuring
AIPROXY_API_TOKENwithout adding the Bearer authentication prefix. Cause: The original credential is used directly to initiate the request, and the authentication header is not completed as required by the model service. - Symptom: The reranker model deployed via docker-compose does not appear in the model list on the FastGPT interface. Cause: The model service port is not exposed to the container network where FastGPT is located, or the full access address of the model is not correctly filled in the configuration.
- Symptom: When the
temperatureparameter is set higher than 0.3, the generated report contains incorrect statutory infectious disease classification entries. Cause: An overly high temperature parameter causes the model output to deviate from the standardized requirements of disease control reports, resulting in random erroneous content.
How to Confirm the Configuration Is Correct
- Initiate a test request containing 10 simulated infectious disease report cards, check that the FastGPT returned interface status code is 200, with no identity verification or format error prompts.
- View the running logs of the model service, confirm that the request initiated by FastGPT carries the correct
AIPROXY_API_TOKENcredential. - Generate a simulated ledger summary result, verify that the output content includes core fields such as statutory infectious disease classification and case count, and the format conforms to the conventional structure of disease control reports.
- Test the incremental data processing flow, confirm that only the case data updated on the current day is included in the summary, and there is no repeated processing of historical data.
The values provided here are common starting points, and should be evaluated against the user's own samples.
Question material comes from public community discussions. Configuration values are common starting points and should be measured against your own samples. Verified on 2026-09-28.