Harness comparisonAI Coding
Kart racer full stage, Claude Code vs two Astra runs — Claude Code (Claude Opus 5.5) vs Claude Code (Claude Sonnet 5.5) vs OpenAI Codex (GPT-6 Astra)
A full Mario Kart-style stage, title screen to finish check, from one prompt: Claude Code with Opus 5.5 and Sonnet 5.5 (left) against two GPT-6 Astra runs in Codex at different thinking levels (center, right).
Each side pairs a coding harness with its own model, so harness and model differences are mixed. More Claude Code vs OpenAI Codex. All harness comparisons.
The post, itemised
- Task
- Kart racer full stage, Claude Code vs two Astra runs
- Category
- Games
- Mode
- One shot
- Posted
- Video length
- 57.8 s (+2 more clips)
- Likes
- 0
- Views
- 119
- Bookmarks
- 0
- Sponsored
- No
Cost & time, as stated in the postAs stated: Claude Code (Opus 5.5 + Sonnet 5.5) 6 h 35 min work time, 7 h 20 min with API time; GPT-6 Astra Codex Ultra 68 min, Extremely high 47 min. No hand-written code.
Related
Posted by @backerd_ceo5 on X. aivs.tv claims no ownership: it hosts a copy for fast playback and links to the post on X. Request removal
