What the data for this category looks like
Data sources for this category include enterprise procurement platforms, public resource trading platforms, supplier databases, contract management systems, and audit issue databases, with daily updates. Daily processed documents cover procurement plans, requirement specifications, bidding documents, bid evaluation records, contracts, acceptance documents, supplier complaint materials, and other related materials. Data fields include procurement entity information, budget amount, procurement method, scoring rules, supplier qualifications, performance and acceptance status, and other relevant details. Each document is linked to corresponding business process nodes, forming complete procurement compliance chain data.
Constraints for Model Access and Configuration
Reference only; verify against the target FastGPT release and deployment documentation.
How to Configure Parameters
| Configuration Item | Recommended Setting | Rationale |
|---|---|---|
maxContext | 8000–12000 token | Covers the complete compliance review context for multiple documents including procurement plans, bidding documents, and bid evaluation records |
temperature | 0.2–0.4 | Ensures the rigor of compliance responses and reduces subjective bias |
quoteMaxToken | 1500–2000 characters | Retains original text references for core compliance documents to meet audit traceability requirements |
toolChoice | auto | Supports automatic invocation of compliance verification tools to check procurement methods and budget basis |
vision | off | The core of this category is text-based compliance consultation, so no image-based business processing is required |
AIPROXY_API_TOKEN | Determined based on actual testing | Ensures valid identity verification for model calls |
Three Common Misconfigurations
- Ollama model calls return empty results. This occurs when
AIPROXY_API_ENDPOINTis not correctly configured to point to the Ollama local service address, or local port mapping is not enabled, preventing network connectivity. - No response is returned during model testing. This occurs when the Ollama service is not started after a system shutdown and restart, or the network configuration of the FastGPT container and Ollama container is inconsistent, preventing communication.
- The
visionparameter is accidentally enabled and an image model is selected. This occurs when the category only handles text-based compliance consultation, but visual capabilities are incorrectly enabled, causing the model to fail to match business requirements.
How to Verify Successful Configuration
- Check that
AIPROXY_API_ENDPOINTandAIPROXY_API_TOKENconfigurations match the deployed model service. - Initiate a single compliance consultation test to verify that the model's returned content links to references from core documents such as procurement plans and bidding documents.
- Test tool invocation to verify procurement method and budget basis checks, confirming that the tool invocation process triggers normally.
- View model call logs to confirm that the token count of returned results does not exceed the
maxContextconfiguration limit.
The values listed here are common starting points and should be evaluated against the user's own sample data.
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.