Harness comparisonAI Coding

Self-playing pinball race, Claude Code vs Codex — Claude Code (Claude Opus 5.5) vs OpenAI Codex (GPT-6.1 Sol)

0:00 / 0:16
Posted byAgentSky@agentsky_dev · on X
View post on X

0 likes96 views

Scoreboard2

  1. Claude Opus 5.5AnthropicClaude Code
    Time
    2.8 min
    Cost
    $0.76
    • Its AI player kept all 3 balls for 40 s
  2. GPT-6.1 SolOpenAIOpenAI Codex
    Time
    83 s
    Cost
    $0.10
    • Its AI player lost all 3 balls by 29 s

Prompt

Self-playing pinball race, Claude Code vs CodexThe post doesn’t quote the prompt wording.

Claude Opus 5.5 in Claude Code and GPT-6.1 Sol in Codex get the same pinball prompt, one run each. In each build an AI player tries to keep three balls alive.

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
Self-playing pinball race, Claude Code vs Codex
Category
Games
Mode
Not stated
Posted
Video length
16.5 s
Likes
0
Views
96
Bookmarks
0
Sponsored
No

NotesNo winner named; the author asks whether faster and cheaper beats the game that plays longer.

Cost & time, as stated in the postAs stated: Claude Code with Claude Opus 5.5 2 min 49 s, $0.76; Codex with GPT-6.1 Sol 1 min 23 s, $0.10.

Related

Posted by @agentsky_dev on X. aivs.tv claims no ownership: it hosts a copy for fast playback and links to the post on X. Request removal

More with these models