aivs.tv

Harness comparisonAI Coding

Supercar drive, three labs' agents on day one — OpenAI Codex (GPT-6.1 Terra) vs Claude Code (Claude Opus 5.5) vs Pi (Gemini 4 Argon)

0:00 / 0:15
Posted byAgentSky@agentsky_dev · on X
View post on X

8 likes635 views

Scoreboard3

  1. GPT-6.1 TerraOpenAIOpenAI Codex
    Time
    3.5 min
    Cost
    $0.12
  2. Claude Opus 5.5AnthropicClaude Code
    Time
    8.3 min
    Cost
    $1.31
  3. Gemini 4 ArgonGooglePi
    Time
    8.7 min
    Cost
    $0.48
    • Second try

Prompt

Supercar drive, three labs' agents on day oneThe post doesn’t quote the prompt wording.

Same supercar prompt on GPT-6.1 Terra's first day, each lab's own agent: Codex with GPT-6.1 Terra, Claude Code with Claude Opus 5.5 and pi with Gemini 4 Argon, labelled on screen.

Each side pairs a coding harness with its own model, so harness and model differences are mixed. More Claude Code vs OpenAI Codex. More OpenAI Codex vs Pi. More Claude Code vs Pi. All harness comparisons.

The post, itemised

Task
Supercar drive, three labs' agents on day one
Mode
Not stated
Posted
Video length
15 s
Likes
8
Views
635
Bookmarks
0
Sponsored
No

NotesThe poster asks which car you would drive.

Cost & time, as stated in the postAs stated: Codex (GPT-6.1 Terra) 3 min 27 s, $0.12; Claude Code (Claude Opus 5.5) 8 min 21 s, $1.31; pi (Gemini 4 Argon) 8 min 41 s, $0.48, second try.

Related

GPT-6.1 TerraClaude Opus 5.5Gemini 4 ArgonCompare all 3 modelsClaude Opus 5.5 vs Gemini 4 ArgonClaude Opus 5.5 vs GPT-6.1 TerraGemini 4 Argon vs GPT-6.1 Terra

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