Expired50% confidenceFactExact time
In the 3D parkour game generation task, GPT 5.5 produced a blank screen in both attempts, scoring 0, while DeepSeek V4 Pro and MiniMax M3 tied with 7 points each.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
5 Models Coding Showdown: Claude, GPT, DeepSeek, M3 — Who's the Best Engineering Tool?
bilibili不正经的前端啊6/8/2026
Related Claims
UnverifiedIn 3D scene iterative correction, DeepSeek V4 resolved the issues in a single round of corrections, whereas GPT-5.5 required multiple rounds of revisions.77% similarUnverifiedIn a 3D smart city monitoring dashboard generation test, GLM-5.1 scored 9.8, DeepSeek V4 scored 9.0, and GPT 5.5 scored 9.5.76% similarUnverifiedGPT-5.5 not only completed the same game rewriting task, but also additionally converted the game into a 3D version.75% similarUnverifiedOn frontend and visual tasks (Three.js animations, SVG graphics), GPT-5.6 models generally only scored six to seven points, while competitor Fable 5 consistently reached nine to ten points.74% similarUnverifiedSomeone used GPT-5.5 to create a fully playable 3D dungeon arena game from scratch in just 6 hours.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/48723API
curl https://kongchang.com/api/v1/knowledge/claims/48723MCP
get_claim(id=48723)