Unverified50% confidenceSolutionExact time
This update adopts a strategy of "advertising capabilities using confirmed model IDs," where only explicitly recognized models will have specific capabilities advertised, and unrecognized model names remain in an unknown state
1
Sources
50%
Confidence
Long-term
Relevance
9/28/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWhen the newly added availableModelsMatch hosting setting is set to exact, availableModels entries will only allow the explicitly named model versions, and new models not on the whitelist will be automatically blocked.69% similarUnverifiedA System Prompt can cause the model to present any specified identity when introducing itself; therefore, asking the model to introduce itself to verify its true identity is technically meaningless.68% similarUnverifiedThe project currently supports only this one specific model, has limited generality, and is positioned as a proof of concept.67% similarUnverifiedThe auxiliary model serves only as a front-end acceleration tool and does not replace the main model in core decision-making; the main model's original response style, semantic logic, and knowledge accuracy remain in their native state.65% similarUnverifiedParameterized policy updates directly modify model weights (e.g., online updating of LoRA fine-tuning layers), offering a higher capability ceiling but introducing the risk of catastrophic forgetting; non-parameterized approaches change behavior by modifying prompts, retrieved content, or tool configurations, offering strong reversibility but limited expressiveness.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/957164API
curl https://kongchang.com/api/v1/knowledge/claims/957164MCP
get_claim(id=957164)