Unverified50% confidenceFactExact time
Today's large models have shown significant improvements in factuality compared to earlier models such as GPT-2.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/25/2026
First Seen
Valid until: 12/24/2026
Sources
Related Entities
Related Claims
VerifiedThe advancement from GPT-3 to GPT-4 was revolutionary, but subsequent improvements have been more about performance optimization on specific tasks rather than a qualitative leap in general capabilities.69% similarUnverifiedIt is considered industry consensus that top international large models (such as Claude and GPT) are an order of magnitude more capable than domestic models.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/949582API
curl https://kongchang.com/api/v1/knowledge/claims/949582MCP
get_claim(id=949582)