Expired90% confidenceEventTime unknown
A Bilibili UP host published a comparative test pitting Claude Sonnet 4.5 against GPT-5 Codex to recreate the soft-body physics driving simulation effect of the 1996 classic game Terep 2 in C++.
1
Sources
90%
Confidence
Medium-term (~90 days)
Relevance
6/2/2026
First Seen
Valid until: 8/31/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedA Chinese tech YouTuber conducted a programming capability comparison test between Claude Sonnet 4.5 and GPT-5 Codex.78% similarUnverifiedIn the C++ test, GPT-5 Codex had a better vehicle model with more pronounced soft-body physics effects, where the vehicle bounced like jelly, more closely resembling the technical demonstration effect of Terep 2.77% similarUnverifiedCopilot CLI's default model is Claude Sonnet 4, and users can switch to models such as GPT-5.73% similarUnverifiedOn the SWE-bench software engineering task benchmark, the task completion rates of Claude 3.5 Sonnet and GPT-4o far exceed those of open-source models.73% similarUnverifiedClaude Sonnet 4 went live before GPT-5.6 Pro.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/37922API
curl https://kongchang.com/api/v1/knowledge/claims/37922MCP
get_claim(id=37922)